Re: [R] Functions in lists or arrays?
t - list() t[[1]] - function(b) b*2 ### NOTE, [[ not [ t [[1]] function (b) b * 2 Bill Venables http://www.cmis.csiro.au/bill.venables/ -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Toby Sent: Tuesday, 21 April 2009 3:51 PM To: r-help@r-project.org Subject: [R] Functions in lists or arrays? I have a problem where I need to have a driver style R program that will extend itself, given more 'source(extra.R)' style lines. IE: easy to modify by other people. The problem becomes when I would like to create an array or list of functions. Each function, possibly hundreds of them, are really created by other programs, generating an *.R file. So for example, if I try: t - list() t[1] - function(b) { b*2 } Error in t[1] - function(b) { : cannot coerce type 'closure' to vector of type 'list' Similar errors for arrays, and anything else I can think of. I'm an R neophite, which likely shows. The only way I seem to be able to do the above, is to generate unique names for each function, and add each name into a list, sort of like this: # Register ourselves models - cbind(models, Nasa_PART_Rules) bounds - cbind(bounds, Nasa_PART_Bounds) Nasa_PART_Rules - NULL Nasa_PART_Bounds - NULL # Rules section Nasa_PART_Rules - rbind(Nasa_PART_Rules, c(Nasa_PART_R1, F)) Nasa_PART_R1 - function(f) { f[,CYCLOMATIC_COMPLEXITY] 8 f[,CYCLOMATIC_COMPLEXITY] = 60 f[,LOC_TOTAL] 73 } Nasa_PART_Bounds - rbind(Nasa_PART_Bounds, c(Nasa_PART_B1)) Nasa_PART_B1 - function(b) { b[CYCLOMATIC_COMPLEXITY,0] - 8 b[CYCLOMATIC_COMPLEXITY,1] - 60 b[LOC_TOTAL,0] - 73 } #... And then using something like this function: # Dispatch a function from its name dispatch - function(f, x) { eval(call(f, x)) } to evaluate each rule over all the data rows: # Read training+validation data dat - read.csv(jm1_data(training+validation).csv) mat - NULL clt - NULL # Evaluate each rule against the dataset for (i in models) { # Get the rules for the model rules - eval(substitute(foo[,1], list(foo=as.name(i cls - eval(substitute(foo[,2], list(foo=as.name(i res - lapply(rules, dispatch, dat) #... Now, this seems way too uggly to me. Can someone give me a hand and/or point me into a more sane direction to explore? One option I have thought of, is to get rid of the *_B?() functions and just fill in a 3 dimensional array using something like: x - NULL dimnames(x) - c(colnames(mat),colnames(dat), c(lbound,ubound)) ... x[RULE_NAME_1, DATA_COL_NAME_1, lbound] - ... ... But I'm not exactly sure how I would construct and/or add onto a global array/etc extra dimnames, as I source each generated *.R file. Anyways, Not sure if I'm making much sense... thanks for any help, -Toby. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] automatic exploration of all possible loglinear models?
Christopher W. Ryan cryan at binghamton.edu writes: Is there a way to automate fitting and assessing loglinear models for several nominal variables . . . something akin to step or drop1 or add1 for linear or logistic regression? Not strictly for loglinear, but glm works with stepAIC. Make sure that in the field you are working this approach is an accepted ritual. Dieter __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sampling in R
__ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sampling in R
__ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Samplin in R
Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=TRUE) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=FALSE) T95-Perm.Cor.Sorted[(nperm+1)*0.05]# 95% treshold value T99-Perm.Cor.Sorted[(nperm+1)*0.01]# 99% treshold value __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sampling in R
(Sorry for multiple posting. Seems to be my msg is not distributed in my previous emails) Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are in the attached file. I want to understand where I am making a mistake. Any comment is deeply appreciated. Kind Regards Seyit Ali Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=TRUE) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=FALSE) T95-Perm.Cor.Sorted[(nperm+1)*0.05]# 95% treshold value T99-Perm.Cor.Sorted[(nperm+1)*0.01]# 99% treshold value __ R-help@r-project.org mailing list
Re: [R] Vizualization of points within sets
Hi, I have a visualization question regarding sets. My problem is as follows: I would define a distance matrix for the points as d_ij = 0 if the points in the same set and 1 if not In case that points can be in more than two sets use an more appropriate distance measure, e.g. Jaccard etc. Then run a nonmetric (two-dimensional) multidimensional scaling (package: MASS, function: isoMDS) to determine the point positions. But I'am not sure that it will be easy to draw the set borders. Yours sincerely Sigbert Klinke __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
Hi r-help-boun...@r-project.org napsal dne 20.04.2009 19:01:46: wolfgang.siewert wolfgang.siewert at gmail.com writes: There is a way around: round(0.7-0.3,1)==0.4 (TRUE) Obviously there is a problem with some combinations of decimal subtractions, that - we have the feeling - shouldt be solved. Oh no, not that one again! This was lecture two in my first computer course in 1968, but it seems to be gone the way of the dodo since than. Maybe that is because of Excel is so widespread now and gives expected results (it probably silently rounds all decimal numbers before calculation). Regards Petr Dietr __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sampling in R: Please read my email from attached text.
(Sorry for multiple posting. Seems to be my msg is not distributed in my previous emails) Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are in the attached file. I want to understand where I am making a mistake. Any comment is deeply appreciated. Kind Regards Seyit Ali Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=TRUE) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=FALSE) T95-Perm.Cor.Sorted[(nperm+1)*0.05]# 95% treshold value T99-Perm.Cor.Sorted[(nperm+1)*0.01]# 99% treshold value __ R-help@r-project.org mailing list
Re: [R] bug when subtracting decimals?
Petr PIKAL petr.pikal at precheza.cz writes: Maybe that is because of Excel is so widespread now and gives expected results (it probably silently rounds all decimal numbers before calculation). Marc Schwartz already reminded me of that one, and it's a good point to explicitly mention in lectures. I suggest to extend R by introducing %==% as being Excellently equal. Dieter __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] R 2.9.0 MASS package
Peter Dalgaard wrote: Tom La Bone wrote: In Windows Xp Pro: R2.8.1 USA(CA1) repository markerSearchPower MASS(VR) MasterBayes R2.9.0 USA(CA1) repository markerSearchPower MasterBayes MASS is not where it used to be. I checked a couple of other repositories in the US and got similar results. Tom But it's in R, no? Anyways, the VR bundle appears to be in the Windows binary area on one CA CRAN mirror but not the other (UCLA doesn't have it, UCB does). Yes, I forgot to enable the recommended packages in my build system directly after release (as it does not make sense to build them before release). The recommended packages from CRAN master should propagate around within a few days. Uwe -p Peter Dalgaard wrote: Tom La Bone wrote: I can't seem to find MASS for the latest version of R. Is it coming or has the name of the package changed? Tom Where did you look? It is in the sources (as part of VR), and also in my (SUSE) test builds. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
r-help-boun...@r-project.org napsal dne 21.04.2009 10:00:06: Petr PIKAL petr.pikal at precheza.cz writes: Maybe that is because of Excel is so widespread now and gives expected results (it probably silently rounds all decimal numbers before calculation). Marc Schwartz already reminded me of that one, and it's a good point to explicitly mention in lectures. I suggest to extend R by introducing %==% as being Excellently equal. It helps but not in all cases (0.7-0.3)-.40 [1] FALSE (0.7-0.3)-.40 [1] TRUE There always could be different issues with not exact representation of decimals. So educated user or internal rounding could help but I am not sure if later is desired. Regards Petr Dieter __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] how to do analysis of PCA with raw data in R software.
Could any body help to how to do analysis of PCA with raw data in R software. I will send you the data then. Regards Abdul Hanan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Question on binomial data
Hi, We have an experiment with pass/fail outcome, and a continuous parameter which may contribute to the outcome. First, we've analyzed it by: p=c(F,T,F,F,F,T,T,T,T,T,T,T,F,T,T,T,T); w=c(53,67,59,59,53,89,72,56,65,63,62,58,59,72,61,68,63); l-glm(p~w,family=binomial) summary(l) Which turned out to be non significant. Then, we thought of comparing the parameters of the two groups (passed vs. failed) t.test(w[which(p)],w[which(!p)],alternative=two.sided) which turned highly significant. I'd appreciate some insight... Thanks, Ehud. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] joined R-today
Friends, I am a newbie to R. Just installed and started with R. I installed netcdf library (netcdf-4.0.tar.gz) and then ncdf package of R from CRAN with the following command. R CMD INSTALL --configure-args=-with-netcdf_incdir=/usr/local/netcdf/include -with-netcdf_libdir=/usr/local/netcdf/lib ncdf_1.6.tar.gz The installation was successful. But when i try to use ncdf inside R, i get the following error. Kindly help me how to resolve the problem. trjfile-system.file('test.netcdf',package='bio3d') trj-read.ncdf(trjfile) Loading required package: ncdf Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in read.ncdf(trjfile) : Please install the ncdf package from CRAN Thanks in advance, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Odp: how to do analysis of PCA with raw data in R software.
Hi r-help-boun...@r-project.org napsal dne 21.04.2009 07:03:45: Could any body help to how to do analysis of PCA with raw data in R software. I will send you the data then. What about you keeping the data and we sending you the software? Or you could download it. Regards Petr Regards Abdul Hanan __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] n stage integer optimisation in R
Hi there, if anyone can please help it would be wonderful! I have a integer programming problem. I have a list of objects, and i need to select them according to certain contraints. In doing so, each object is given a value, and this value is what needs to be maximised. This is not the problem, i use a simple integer program(using lpSolve package), with a constraint matrix, setting all.bin = TRUE, and the 1's will indicate that the object must be selected, a zero indicates that it shouldn't be selected. So selecting the objects at this stage is fine, but here is my big problem: There are multiple stages and each object is given different score a each stage. I now need to optimise the total (over all 10 stages) knowing that i can only make x changes to the set objects chosen at the first stage. does anyone know of a package i can use that can do this for me? If the problem is not clear, please let me know so i can try to rephrase it Thanx -- View this message in context: http://www.nabble.com/n-stage-integer-optimisation-in-R-tp23151632p23151632.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Hi, Check out the FAQ on the home page about installing packages. You need to 1. Install the package from a local mirror, you can do this from the drop down menu (Install packages) 2. then call the package using library(name of the package) It helps to know where your R library directory is on your computer. HTH Simon. - Original Message - From: Bala subramanian bala.biophys...@gmail.com To: r-help@r-project.org Sent: Tuesday, April 21, 2009 9:34 AM Subject: [R] joined R-today Friends, I am a newbie to R. Just installed and started with R. I installed netcdf library (netcdf-4.0.tar.gz) and then ncdf package of R from CRAN with the following command. R CMD INSTALL --configure-args=-with-netcdf_incdir=/usr/local/netcdf/include -with-netcdf_libdir=/usr/local/netcdf/lib ncdf_1.6.tar.gz The installation was successful. But when i try to use ncdf inside R, i get the following error. Kindly help me how to resolve the problem. trjfile-system.file('test.netcdf',package='bio3d') trj-read.ncdf(trjfile) Loading required package: ncdf Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in read.ncdf(trjfile) : Please install the ncdf package from CRAN Thanks in advance, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] line wrap in R console under windows
I would like the R console to wrap lines at 80 cols. It does not do so, even though I have used the Rgui Configuration Editor to set the Console cols at 80 and the Pager cols at 80. Please tell me how to set it up so I have word wrap. Just to be clear: in older/other R versions, console input is wrapped at 80 cols and so you have something that looks like blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah Instead of what I see, which is blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah$ and I need to scroll sideways to see the rest of the stuff past $ I checked the archives and saw no postings on this. Thanks very much for any help. Bill __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Hi, Yes ncdf needs netcdf, netcdf is already installed in my fedora10. I reinstalled ncdf with the following option install.packages('ncdf',dep=TRUE) library(ncdf) library(ncdf) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Again the same error, Bala On Tue, Apr 21, 2009 at 12:12 PM, Liviu Andronic landronim...@gmail.comwrote: On Tue, Apr 21, 2009 at 11:57 AM, Bala subramanian bala.biophys...@gmail.com wrote: Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. Does ncdf depend on netcdf [1]? If so, perhaps it is missing, or is not up-to-date. Also, it's a good habit to install.packages('ncdf', dep=TRUE). Regards, Liviu [1] http://www.unidata.ucar.edu/software/netcdf/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Sampling in R
Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are below Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=TRUE) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=FALSE) T95-Perm.Cor.Sorted[(nperm+1)*0.05]# 95% treshold value T99-Perm.Cor.Sorted[(nperm+1)*0.01]# 99% treshold value I want to understand where I am making a mistake. Any comment is deeply appreciated. Kind Regards Seyit Ali -- Dr. Seyit Ali KAYIS Selcuk University Faculty of Agriculture Kampus, Konya, TURKEY
[R] Sampling in R
Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are in the attached file. I want to understand where I am making a mistake. Any comment is deeply appreciated. Kind Regards Seyit Ali -- Dr. Seyit Ali KAYIS Selcuk University Faculty of Agriculture Kampus, Konya, TURKEY s_a_ka...@yahoo.com,s_a_ka...@hotmail.com Tell: +90 332 223 2830 Mobile: +90 535 587 1139 Fax: +90 332 241 0108 Greetings from Konya, TURKEY http://www.ziraat.selcuk.edu.tr/skayis/ -- _ No-one wants to be lonely this Autumn Find someone to snuggle up with Fchannel%2Findex%2Easpx%3Ftrackingid%3D1048628_t=773568480_r=nzWINDOWSliveMAILemailTAGLINES_m=EXTXvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar),
Re: [R] joined R-today
Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. *install.packages('ncdf')* Warning in install.packages(ncdf) : argument 'lib' is missing: using '/usr/lib/R/library' --- Please select a CRAN mirror for use in this session --- Loading Tcl/Tk interface ... done trying URL ' http://rm.mirror.garr.it/mirrors/CRAN/src/contrib/ncdf_1.6.tar.gz' Content type 'application/x-gzip' length 72533 bytes (70 Kb) opened URL == downloaded 70 Kb * Installing *source* package 'ncdf' ... checking for gcc... gcc -m32 -std=gnu99 checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc -m32 -std=gnu99 accepts -g... yes checking for gcc -m32 -std=gnu99 option to accept ANSI C... none needed checking how to run the C preprocessor... gcc -m32 -std=gnu99 -E checking for egrep... grep -E checking for ANSI C header files... yes checking for sys/types.h... yes checking for sys/stat.h... yes checking for stdlib.h... yes checking for string.h... yes checking for memory.h... yes checking for strings.h... yes checking for inttypes.h... yes checking for stdint.h... yes checking for unistd.h... yes checking netcdf.h usability... yes checking netcdf.h presence... yes checking for netcdf.h... yes Found netcdf.h in: . checking for nc_open in -lnetcdf... yes Found netcdf library file libnetcdf.a in directory . configure: creating ./config.status config.status: creating R/load.R config.status: creating src/Makevars ** libs gcc -m32 -std=gnu99 -I/usr/include/R -I. -I/usr/local/include-fpic -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i386 -mtune=generic -fasynchronous-unwind-tables -c ncdf2.c -o ncdf2.o gcc -m32 -std=gnu99 -I/usr/include/R -I. -I/usr/local/include-fpic -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i386 -mtune=generic -fasynchronous-unwind-tables -c ncdf3.c -o ncdf3.o ncdf3.c: In function R_nc_get_vara_charvarid: ncdf3.c:221: warning: assignment discards qualifiers from pointer target type ncdf3.c: In function R_nc_get_vara_numvarid: ncdf3.c:267: warning: assignment discards qualifiers from pointer target type ncdf3.c:249: warning: rv_data may be used uninitialized in this function gcc -m32 -std=gnu99 -I/usr/include/R -I. -I/usr/local/include-fpic -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i386 -mtune=generic -fasynchronous-unwind-tables -c ncdf.c -o ncdf.o ncdf.c: In function R_nc_ttc_to_nctype: ncdf.c:424: warning: implicit declaration of function exit ncdf.c:424: warning: incompatible implicit declaration of built-in function exit gcc -m32 -std=gnu99 -shared -L/usr/local/lib -o ncdf.so ncdf2.o ncdf3.o ncdf.o -L. -lnetcdf -L/usr/lib/R/lib -lR ** R ** preparing package for lazy loading ** help Building/Updating help pages for package 'ncdf' Formats: text html latex example ancdf texthtmllatex aput.var.ncdf texthtmllatex example Note: unmatched right brace in file 'att.get.ncdf.Rd' on or after line 6 att.get.ncdf texthtmllatex example att.put.ncdf texthtmllatex example close.ncdftexthtmllatex example create.ncdf texthtmllatex example dim.def.ncdf texthtmllatex example enddef.ncdf texthtmllatex example get.var.ncdf texthtmllatex example ncdf-internal texthtmllatex open.ncdf texthtmllatex example print.ncdftexthtmllatex example redef.ncdftexthtmllatex example set.missval.ncdf texthtmllatex example sync.ncdf texthtmllatex example var.add.ncdf texthtmllatex example var.def.ncdf texthtmllatex example Note: unmatched right brace in file 'version.ncdf.Rd' on or after line 6 Note: removing empty section \arguments in file 'version.ncdf.Rd' version.ncdf texthtmllatex ** building package indices ... ** DONE (ncdf)* The downloaded packages are in /tmp/RtmpKrf8Vc/downloaded_packages Updating HTML index of packages in '.Library' *library(ncdf)* Error in dyn.load(file, DLLpath = DLLpath, ...) : unable
[R] search through a matrix
Hi. I have a 925 by 925 correlation matrix corM. I want to identify all variables that have correlation greater than 0.9. Can anyone suggest an R way of doing this? Thank you. -- View this message in context: http://www.nabble.com/search-through-a-matrix-tp23153538p23153538.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
On 21/04/2009 3:48 AM, Petr PIKAL wrote: Hi r-help-boun...@r-project.org napsal dne 20.04.2009 19:01:46: wolfgang.siewert wolfgang.siewert at gmail.com writes: There is a way around: round(0.7-0.3,1)==0.4 (TRUE) Obviously there is a problem with some combinations of decimal subtractions, that - we have the feeling - shouldt be solved. Oh no, not that one again! This was lecture two in my first computer course in 1968, but it seems to be gone the way of the dodo since than. Maybe that is because of Excel is so widespread now and gives expected results (it probably silently rounds all decimal numbers before calculation). I don't have Excel, but I expect OpenOffice duplicates its bugs pretty well. And in OpenOffice I see all sorts of bugs due to this, e.g. examples where x = y and y = z but x != z, cases where I can calculate a number like 1 + 4.e-15 and get something different from 1, but if I enter it directly as 1.004, it gets changed to 1. So it only gives expected results in some tests, not others. Duncan Murdoch __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Odp: Sampling in R
Hi r-help-boun...@r-project.org napsal dne 21.04.2009 12:25:01: Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are below Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185, 0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0. 1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492, 0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729, 0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469, 0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636, 0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091, 0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143, 0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158, 0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0. 5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182, 0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0. 2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0. 6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0. 4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58, 0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406, 0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0. 4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0. 4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0. 367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0. 5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } AFAICU you do not sample your data you shuffle them. Then you compute cor with shuffled data (X and Y are shuffled independently) which results in low correlation (it is like shuffling cards). Maybe you could use smaller size and sample not original data but a vector of indices perm.cor-rep(NA, 4) for (iperm in 1:nperm) { ind - sample(1:length(Xvar), size = 100, replace=FALSE) perm.cor[iperm] - cor(Xvar[ind], Yvar[ind]) perm.cor } max(perm.cor) hist(perm.cor) The result seems to be quite reasonable. Regards Petr print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor,
Re: [R] search through a matrix
Uwe Ligges wrote: onyourmark wrote: Hi. I have a 925 by 925 correlation matrix corM. I want to identify all variables that have correlation greater than 0.9. Can anyone suggest an R way of doing this? Thank you. Example: # prepare example data: x - matrix((1:25) / 25, 5, 5) rownames(x) - letters[1:5] colnames(x) - letters[1:5] # solution cbind(rownames(x)[row(x)[x 0.9]], colnames(x)[col(x)[x 0.9]]) or even which(x 0.9, arr.ind = TRUE) Best, Dimitris Uwe Ligges __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitris Rizopoulos Assistant Professor Department of Biostatistics Erasmus University Medical Center Address: PO Box 2040, 3000 CA Rotterdam, the Netherlands Tel: +31/(0)10/7043478 Fax: +31/(0)10/7043014 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] line wrap in R console under windows
On 21/04/2009 6:03 AM, William Simpson wrote: I would like the R console to wrap lines at 80 cols. It does not do so, even though I have used the Rgui Configuration Editor to set the Console cols at 80 and the Pager cols at 80. Please tell me how to set it up so I have word wrap. Just to be clear: in older/other R versions, console input is wrapped at 80 cols and so you have something that looks like blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah Instead of what I see, which is blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah$ and I need to scroll sideways to see the rest of the stuff past $ Which version are you talking about? I don't recall Rgui ever wrapping input. The config settings affect output. Maybe you're thinking of Rterm? Duncan Murdoch I checked the archives and saw no postings on this. Thanks very much for any help. Bill __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Package or packages for randomization in a clinical trial
Can anyone recommend a package that can be used to randomize subjects? I am looking for a generalized package, or several packages that can accomplish unrestricted randomization (i.e. simple random assignment) restricted randomization including stratified randomization, blocked randomization, and adaptive randomization. Thanks, John John David Sorkin M.D., Ph.D. Chief, Biostatistics and Informatics University of Maryland School of Medicine Division of Gerontology Baltimore VA Medical Center 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 (Phone) 410-605-7119 (Fax) 410-605-7913 (Please call phone number above prior to faxing) Confidentiality Statement: This email message, including any attachments, is for th...{{dropped:6}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] How to compare parameters of non linear fitting curves
Hi, I'm using a non linear model to fit experimental survival curves. This model describes the fraction of still active experiments as a function of time t as follows: f(t)=(1+exp(-etaD*cD)) / (1+exp(etaD(t-cD))) Moreover, when experiments are still active, they may change of state (from 0 to 1). But they may fall inactive before changing their state (their state still equals 0). The survival curve of state may also be fitted with the following model: f(A)=(1+exp(-eta1*c1)) / (1+exp(eta1(t-c1))) * (1+exp(-etaD*cD)) / (1+exp(etaD(t-cD))) I estimate with nlm 1°) values of etaD and cD parameters and 2°) inject them as constant in the function to be minimized by nlm to estimate values of eta1 and c1. I perform these estimations for two different experimental conditions that both have their values of etaD,eta1, cD and c1. I would like to know if there is any statistical method to compare the estimated values of parameters of the two distributions ? And wether it's the case, how to perform it in R ? Hope I'm clear enough for getting help, Etienne --- Etienne Toffin, PhD Student Unit of Social Ecology Université Libre de Bruxelles, CP 231 Boulevard du Triomphe B-1050 Brussels Belgium Tel: +32(0)2/650.55.30 Fax: +32(0)2/650.57.67 http://www.ulb.ac.be/sciences/use/toffin.html __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] search through a matrix
onyourmark wrote: Hi. I have a 925 by 925 correlation matrix corM. I want to identify all variables that have correlation greater than 0.9. Can anyone suggest an R way of doing this? Thank you. Example: # prepare example data: x - matrix((1:25) / 25, 5, 5) rownames(x) - letters[1:5] colnames(x) - letters[1:5] # solution cbind(rownames(x)[row(x)[x 0.9]], colnames(x)[col(x)[x 0.9]]) Uwe Ligges __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sampling in R
When you shuffle the observations independently, you are performing a permutation test (though for this you only need to shuffle one side of the pairs). When you sort the observations you are doing something ridiculous that has no statistical meaning that I know. I'm not very familiar with bootstrap CI's, but I think the idea is to sample the PAIRS of data WITH replacement: http://lmgtfy.com/?q=bootstrap+correlation (first link is to a good overview by David Howell) On Tue, Apr 21, 2009 at 7:25 AM, Seyit Ali Kayis s_a_ka...@hotmail.com wrote: Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are below Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0)
Re: [R] Sampling in R: Please read my email from attached text.
You need to sample pairs rather than sampling individually within Xvar and Yvar. You also generally sample with replacement. If you sample without replacement for the length of the data, then you just get the same set. On Apr 21, 2009, at 3:54 AM, skayis selcuk wrote: Data_and_Sampling_codes.txt David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Question on binomial data
You should review your course material on interpreting general linear models. The criterion you have chosen for significance (looking at p values for indivdiual coefficients) is not the recommended one. Seek out the section that discusses the proper method for using deviance estimates for comparing nested models. -- David Winsemius On Apr 21, 2009, at 4:32 AM, ehud cohen wrote: Hi, We have an experiment with pass/fail outcome, and a continuous parameter which may contribute to the outcome. First, we've analyzed it by: p=c(F,T,F,F,F,T,T,T,T,T,T,T,F,T,T,T,T); w=c(53,67,59,59,53,89,72,56,65,63,62,58,59,72,61,68,63); l-glm(p~w,family=binomial) summary(l) Which turned out to be non significant. Then, we thought of comparing the parameters of the two groups (passed vs. failed) t.test(w[which(p)],w[which(!p)],alternative=two.sided) which turned highly significant. I'd appreciate some insight... David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Try to install from scratch from a different mirror. Dimitri On Tue, Apr 21, 2009 at 6:21 AM, Bala subramanian bala.biophys...@gmail.com wrote: Hi, Yes ncdf needs netcdf, netcdf is already installed in my fedora10. I reinstalled ncdf with the following option install.packages('ncdf',dep=TRUE) library(ncdf) library(ncdf) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Again the same error, Bala On Tue, Apr 21, 2009 at 12:12 PM, Liviu Andronic landronim...@gmail.comwrote: On Tue, Apr 21, 2009 at 11:57 AM, Bala subramanian bala.biophys...@gmail.com wrote: Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. Does ncdf depend on netcdf [1]? If so, perhaps it is missing, or is not up-to-date. Also, it's a good habit to install.packages('ncdf', dep=TRUE). Regards, Liviu [1] http://www.unidata.ucar.edu/software/netcdf/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Random Forests: Question about R^2
Just one small correction: in #3 it should be squared residuals. Yes, the function returns a vector of r^2 with length=ntree, with the k-th element being the r^2 for the forest consisting of the first k trees. Cheers, Andy From: Dimitri Liakhovitski I would like to summarize. Would you please confirm that my summary is correct? Thank you very much! Determining R^2 in Random Forests (for a Regression Forest): 1. For each individual case, record a mean prediction on the dependent variable y across all trees for which the case is OOB (Out-of-Bag); 2. For each individual case, calculate a residual: residual = observed y - mean predicted y (from step 1) 3. Calculate mean square residual MSE: MSE = sum of all individual residuals (from step 2) / n 4. Because MSE/var(y) represents the proportion of y variance that is due to error, then R^2 = 1 - MSE/var(y). If it's correct, my last question would be: I am getting as many R^2 as the number of trees because each time the residuals are recalculated using all trees built so far, correct? Thank you very much! Dimitri On Mon, Apr 13, 2009 at 6:22 PM, Liaw, Andy andy_l...@merck.com wrote: Apologies: that should have been sum(residual^2)! -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Monday, April 13, 2009 4:35 PM To: Liaw, Andy Cc: R-Help List Subject: Re: [R] Random Forests: Question about R^2 Andy, thank you very much! One clarification question: If MSE = sum(residuals) / n, then in the formula (1 - mse / Var(y)) - shouldn't one square mse before dividing by variance? Dimitri On Mon, Apr 13, 2009 at 10:52 AM, Liaw, Andy andy_l...@merck.com wrote: MSE is the mean squared residuals. For the training data, the OOB estimate is used (i.e., residual = data - OOB prediction, MSE = sum(residuals) / n, OOB prediction is the mean of predictions from all trees for which the case is OOB). It is _not_ the average OOB MSE of trees in the forest. I hope there's no question about how the pseudo R^2 is computed on a test set? If you understand how that's done, I assume the confusion is only how the OOB MSE is formed. Best, Andy From: Dimitri Liakhovitski Dear Random Forests gurus, I have a question about R^2 provided by randomForest (for regression). I don't succeed in finding this information. In the help file for randomForest under Value it says: rsq: (regression only) - pseudo R-squared'': 1 - mse / Var(y). Could someone please explain in somewhat more detail how exactly R^2 is calculated? Is mse mean squared error for prediction? Is mse an average of mse's for all trees run on out-of-bag holdout samples? In other words - is this R^2 based on out-of-bag samples? Thank you very much for clarification! -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Notice: This e-mail message, together with any attachments, contains information of Merck Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp Dohme or MSD and in Japan, as Banyu - direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com Notice: This e-mail message, together with any attachments, contains information of Merck Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp Dohme or MSD and in Japan, as Banyu - direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. -- Dimitri Liakhovitski MarketTools, Inc.
Re: [R] Random Forests: Predictor importance for Regression Trees
Yes, you've got it! Cheers, Andy From: Behalf Of Dimitri Hello! I think I am relatively clear on how predictor importance (the first one) is calculated by Random Forests for a Classification tree: Importance of predictor P1 when the response variable is categorical: 1. For out-of-bag (oob) cases, randomly permute their values on predictor P1 and then put them down the tree 2. For a given tree, subtract the number of votes for the correct class in the predictor-P1-permuted oob dataset from the number of votes for the correct class in the untouched oob dataset: if P1 is important, this number will be large. 3. The average of this number over all trees in the forest is the raw importance score for predictor P1. I am wondering what step 2 above looks like if the response variable is continous and not categorical, in other words - for a Regression tree. Could you please correct if what I wrote below is wrong? Thank you very much! Importance of predictor P1 when the response variable is continous: 1. For out-of-bag (oob) cases, randomly permute their values on predictor P1 and then put them down the tree 2. For a given tree, calculate mean squared deviation of observed y minus predicted y for (a) the untouched oob dataset and for (b) the predictor-P1-permuted oob dataset. Subtract (a) from (b). 3. The average of this number over all trees in the forest is the raw importance score for predictor P1. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Notice: This e-mail message, together with any attachme...{{dropped:12}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Hi, I tried to install from four different mirrors. The result is the same. I dnt understand what the following error means. /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' I even tried to change the permission of ncdf.so but still i am not able to load the library. Bala On Tue, Apr 21, 2009 at 2:03 PM, Dimitri Liakhovitski ld7...@gmail.comwrote: Try to install from scratch from a different mirror. Dimitri On Tue, Apr 21, 2009 at 6:21 AM, Bala subramanian bala.biophys...@gmail.com wrote: Hi, Yes ncdf needs netcdf, netcdf is already installed in my fedora10. I reinstalled ncdf with the following option install.packages('ncdf',dep=TRUE) library(ncdf) library(ncdf) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Again the same error, Bala On Tue, Apr 21, 2009 at 12:12 PM, Liviu Andronic landronim...@gmail.com wrote: On Tue, Apr 21, 2009 at 11:57 AM, Bala subramanian bala.biophys...@gmail.com wrote: Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. Does ncdf depend on netcdf [1]? If so, perhaps it is missing, or is not up-to-date. Also, it's a good habit to install.packages('ncdf', dep=TRUE). Regards, Liviu [1] http://www.unidata.ucar.edu/software/netcdf/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sampling in R
Dear Seyit, You might consider the boot package in this situation. Here is an example: require(boot) DF-data.frame(Xvar,Yvar) temp - boot(DF, function(DF,d){ S - DF[d,] cor(S$Xvar,S$Yvar) }, R = 5000) temp$t0 # [1] 0.617422 max(temp$t) # [1] 0.7783784 hist(temp$t) Once you load the boot package, take a look at ?boot. HTH, Jorge On Tue, Apr 21, 2009 at 4:53 AM, Seyit Ali Kayis s_a_ka...@hotmail.comwrote: Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are in the attached file. I want to understand where I am making a mistake. Any comment is deeply appreciated. Kind Regards Seyit Ali -- Dr. Seyit Ali KAYIS Selcuk University Faculty of Agriculture Kampus, Konya, TURKEY s_a_ka...@yahoo.com,s_a_ka...@hotmail.com Tell: +90 332 223 2830 Mobile: +90 535 587 1139 Fax: +90 332 241 0108 Greetings from Konya, TURKEY http://www.ziraat.selcuk.edu.tr/skayis/ -- _ No-one wants to be lonely this Autumn Find someone to snuggle up with Fchannel%2Findex%2Easpx%3Ftrackingid%3D1048628_t=773568480_r=nzWINDOWSliveMAILemailTAGLINES_m=EXT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Hi, In install.packages, there is something called *lib* which says, lib: character vector giving the library directories where to install the packages. Recycled as needed. If missing, defaults to '.libPaths()[1]'. If i have to give the lib option as an arguement, what should be the path for the installation, should it be */usr/local/lib* or */usr/lib/R/lib* or * /usr/lib/R/library/ncdf/libs/* Bala On Tue, Apr 21, 2009 at 2:14 PM, Bala subramanian bala.biophys...@gmail.com wrote: Hi, I tried to install from four different mirrors. The result is the same. I dnt understand what the following error means. /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' I even tried to change the permission of ncdf.so but still i am not able to load the library. Bala On Tue, Apr 21, 2009 at 2:03 PM, Dimitri Liakhovitski ld7...@gmail.comwrote: Try to install from scratch from a different mirror. Dimitri On Tue, Apr 21, 2009 at 6:21 AM, Bala subramanian bala.biophys...@gmail.com wrote: Hi, Yes ncdf needs netcdf, netcdf is already installed in my fedora10. I reinstalled ncdf with the following option install.packages('ncdf',dep=TRUE) library(ncdf) library(ncdf) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Again the same error, Bala On Tue, Apr 21, 2009 at 12:12 PM, Liviu Andronic landronim...@gmail.comwrote: On Tue, Apr 21, 2009 at 11:57 AM, Bala subramanian bala.biophys...@gmail.com wrote: Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. Does ncdf depend on netcdf [1]? If so, perhaps it is missing, or is not up-to-date. Also, it's a good habit to install.packages('ncdf', dep=TRUE). Regards, Liviu [1] http://www.unidata.ucar.edu/software/netcdf/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] explicit documentation
Duncan Murdoch wrote: ronggui wrote: It is always unfair to complain about volunteer work, and what you should do is to make contributions. I only half agree with this: ... and i half agree with this (thus quarter agree with that). i think it's pretty fair to complain about unclear documentation *without* proposing a rewrite. after all, one may be complaining because one is unable to understand from the existing docs what the intended and to-be-documented behaviour is, and this is hardly a position from which to propose concrete modifications. but a message that a help page is unclear can still be useful -- it should, however, clearly explain why the page is not considered clear. vQ I think that it's fair to complain, as long as you make contributions. With documentation, if you don't think it's clear, as part of your complaint you should rewrite it in a way that is clear: then the author can understand what you found unclear about it. (It may well happen that your revision isn't accepted, because it may be less clear than the original in the author's opinion, or it may be incorrect: but at least it saves the author the time of trying to figure out what you'd prefer.) Duncan Murdoch 2009/4/20 Rolf Turner r.tur...@auckland.ac.nz: On 19/04/2009, at 8:59 PM, Patrick Burns wrote: Rolf Turner wrote: On 17/04/2009, at 10:21 PM, Duncan Murdoch wrote: Benjamin Tyner wrote: Many thanks Duncan. Perhaps this merits a more explicit note in the documentation? The quote I gave is from the documentation. How could it be more explicit? This is unfortunately typical of the attitude of R-core people toward the documentation. ``It's clear.'' they say. ``It's explicit.'' Clear and explicit once you *know* what it's saying. Not before, but. I think this unfairly blames R-core for being human. Why is this unfair? R-core is supposed to be superhuman! :-) cheers, Rolf __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Fix/Edit Error
Hi, I am wondering if anyone can help me with the following error. I am using R on a Linux machine and wish to use either the edit or fix function to look at the data. However, we I try to run either of these commands I get the following error Error in data entry(datalist, modes) : invalid device In addition: Warning message: In edit.data.frame(get(subx, envir = parent), title = subx, ...) : unable to create fontset -*-fixed-medium-r-*-*-*-120-*-*-*-*-*-* http://www.google.co.uk/search?hl=enei=h6ntSbXfHMeTjAe727ANsa=Xoi=sp ellresnum=0ct=resultcd=1q=Error+in+data+entry(datalist,+modes)+%3A+i nvalid+device+In+addition%3A+Warning+message%3A+In+edit.data.frame(get(s ubx,+envir+%3D+parent),+title+%3D+subx,+...)+%3A+unable+to+create+fontse t+-*-fixed-medium-r-*-*-*-120-*-*-*-*-*-*spell=1 Does anyone know how I can get around this so I can look at the data? Many thanks in advance, [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
On Tue, Apr 21, 2009 at 11:57 AM, Bala subramanian bala.biophys...@gmail.com wrote: Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. Does ncdf depend on netcdf [1]? If so, perhaps it is missing, or is not up-to-date. Also, it's a good habit to install.packages('ncdf', dep=TRUE). Regards, Liviu [1] http://www.unidata.ucar.edu/software/netcdf/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] lme for between-within anova
I have the following between-within anova: aovn1 - aov(amplitude ~ stereo*site*stimulus + Error(subject/(site*stimulus)), stereon1) This works fine. BUT I need to do Tukey HSD multiple comparisons, and the aov() approach won't work. So I am trying the method posted on r-help: lmen1 - lme(amplitude ~ stereo*site*stimulus, random = ~1|subject/(site*stimulus), stereon1) This doesn't work: lmen1 - lme(amplitude ~ stereo*site*stimulus, random = ~1|subject/(site*stimulus), stereon1) Error in getGroups.data.frame(dataMix, groups) : Invalid formula for groups Please tell me what to do to get the lme() fit to work. After I get lmen1, I will do: anova(lmen1) summary(lmen1) summary(glht(lmen1, linfct=mcp(V=Tukey))) Please tell me if that sounds right. Thanks very much for any help! Bill = This is the example I am following (with modifications for my expt design) You want to use lme() in package nlme, then glht() in the multcomp package. This will give you multiplicity adjusted p-values and confidence intervals. ## Example require(MASS) ## for oats data set require(nlme) ## for lme() require(multcomp) ## for multiple comparison stuff Aov.mod - aov(Y ~ N + V + Error(B/V), data = oats) Lme.mod - lme(Y ~ N + V, random = ~1 | B/V, data = oats) summary(Aov.mod) anova(Lme.mod) summary(Lme.mod) summary(glht(Lme.mod, linfct=mcp(V=Tukey))) __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] rmysql fetch issues
I am trying to fetch data using RMySQL fetch method with n=100 in this loop: lquery = paste(query, limit 10) input = dbGetQuery(con1,lquery) res = dbSendQuery(con1,query) completed = FALSE count = 0 while (completed != TRUE) { count = count + 1 batch = fetch(res, n = 100) print (paste(batch,count)) input = merge(input,batch,all=TRUE) print (paste(merge,count)) completed = dbHasCompleted(res) } But after a while, I get this error: ... [1] batch 211 [1] merge 211 [1] batch 212 Error in rep.int(rep.int(seq_len(nx), rep.int(rep.fac, nx)), orep) : invalid 'times' value In addition: Warning messages: 1: In mysqlFetch(res, n, ...) : RS-DBI driver warning: (error while fetching rows) 2: In mysqlFetch(res, n, ...) : RS-DBI driver warning: (error while fetching rows) What could be causing this? Is there any other way to do this? [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Sampling in R
Seyit Ali Kayis wrote: Dear R users, I need to do sampling without replacement (bootstraps). I have two variables (Xvar, Yvar). I have a correlation from original data set cor(Xvar, Yvar)=0.6174221. I am doing 5 sampling, and in each sampling calculating correlations, saving, sorting and getting 95% cutt off point (0.1351877). I am getting maximum value as 0.3507219 (much smaller than correlation of my original data). I repeated the sampling a couple of time and none of them produced a correlation coefficient higher than my original data set. However, if I sort out my Xvar and Yvar and obtain correlation it is 0.9657125 which is much higher than correlation for my original data. I am doing sampling in another program and getting at least 1% higher correlation than mine. Now I am getting confused with sampling(random data) in R. My data and codes for the scenario above are below Xvar-c(0.1818182,0.5384615,0.5535714,0.4680851,0.4545455,0.4385965,0.5185185,0.4035088,0.4901961,0.3650794,0.462963,0.4,0.56,0.3965517,0.4909091, 0.4716981,0.4310345,0.2,0.1509434,0.2647059,0.173913,0.1914894,0.1914894,0.1489362,0.1363636,0.2244898,0.2325581,0.133,0.1818182,0.1702128, 0.2173913,0.2380952,0.1632653,0.5614035,0.3396226,0.4909091,0.3770492,0.5,0.5185185,0.5,0.467,0.4464286,0.362069,0.4285714,0.4561404, 0.4736842,0.4545455,0.417,0.4181818,0.4590164,0.517,0.5423729,0.483,0.5454545,0.4393939,0.5172414,0.4098361,0.4745763,0.4754098, 0.517,0.5,0.4603175,0.42,0.4038462,0.4897959,0.3148148,0.3673469,0.4,0.458,0.3877551,0.4375,0.4117647,0.4313725,0.533,0.3962264, 0.3548387,0.5272727,0.4137931,0.3928571,0.467,0.4210526,0.4363636,0.4545455,0.4310345,0.4237288,0.4814815,0.4912281,0.433,0.4,0.4285714, 0.4516129,0.5090909,0.4464286,0.4642857,0.417,0.4098361,0.4909091,0.3809524,0.5272727,0.4814815,0.5254237,0.627451,0.5,0.5471698,0.5454545, 0.5925926,0.5769231,0.5818182,0.444,0.4915254,0.4727273,0.4107143,0.4285714,0.4310345,0.4237288,0.4285714,0.440678,0.4237288,0.4807692, 0.4150943,0.4615385,0.4107143,0.4814815,0.4074074,0.4210526,0.5263158,0.440678,0.4576271,0.5344828,0.5,0.5636364,0.4677419,0.5,0.5192308, 0.4642857,0.5090909,0.58,0.4482759,0.5098039,0.4035088,0.4210526,0.5098039,0.4385965,0.5283019,0.5471698,0.625,0.4310345,0.4912281,0.5283019, 0.4576271,0.5471698,0.4745763,0.4821429) Yvar-c(0.2553191,0.4107143,0.5660377,0.389,0.3606557,0.2898551,0.3818182,0.4,0.4,0.3278689,0.2903226,0.4074074,0.4181818,0.3,0.2238806,0.3728814, 0.3709677,0.2307692,0.2830189,0.2244898,0.2142857,0.2131148,0.22,0.2258065,0.2321429,0.2,0.2264151,0.22,0.2115385,0.2459016,0.117,0.1785714, 0.2068966,0.6,0.4285714,0.3134328,0.4461538,0.3965517,0.4769231,0.6181818,0.4827586,0.3709677,0.3965517,0.4821429,0.4545455,0.359375,0.4576271, 0.4516129,0.5272727,0.4603175,0.4,0.4912281,0.5384615,0.5,0.4516129,0.4126984,0.4655172,0.5263158,0.4925373,0.358209,0.4285714,0.4920635, 0.4482759,0.3235294,0.4,0.4375,0.440678,0.3898305,0.35,0.4528302,0.58,0.4153846,0.3174603,0.5185185,0.3870968,0.2894737,0.3709677,0.369863, 0.3676471,0.3636364,0.3088235,0.328125,0.4032258,0.4084507,0.3188406,0.3636364,0.3823529,0.2816901,0.472,0.5,0.3521127,0.4393939,0.3787879, 0.453125,0.4324324,0.4057971,0.4545455,0.4492754,0.5,0.4098361,0.4067797,0.367,0.3928571,0.4285714,0.5,0.2923077,0.4561404,0.45,0.5538462, 0.4626866,0.4057971,0.3676471,0.5322581,0.5428571,0.375,0.4411765,0.4571429,0.4,0.3846154,0.3870968,0.4915254,0.530303,0.4375,0.4918033,0.4179104, 0.4032258,0.3606557,0.5178571,0.4848485,0.390625,0.375,0.4375,0.367,0.4,0.4477612,0.2571429,0.4032258,0.3382353,0.4814815,0.4090909,0.3548387, 0.4821429,0.5,0.557377,0.433,0.5454545,0.4590164,0.3943662,0.5076923,0.5,0.3283582,0.3676471,0.559322) my.cor-cor(Xvar, Yvar) print(my.cor) nperm-4 Perm.Cor-NULL for (iperm in 1:nperm) { XvarNew-sample(Xvar, size=length(Xvar), replace=FALSE) YvarNew-sample(Yvar, size=length(Yvar), replace=FALSE) perm.cor-cor(XvarNew, YvarNew) Perm.Cor-c(Perm.Cor, perm.cor) } print(max(Perm.Cor)) XvarSorted-sort(Xvar, decreasing=TRUE) YvarSorted-sort(Yvar, decreasing=TRUE) max.cor-cor(XvarSorted, YvarSorted) print(max.cor) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=TRUE) if(mat.cor0) Perm.Cor.Sorted-sort(Perm.Cor, decreasing=FALSE) T95-Perm.Cor.Sorted[(nperm+1)*0.05]# 95% treshold value T99-Perm.Cor.Sorted[(nperm+1)*0.01]# 99% treshold value I want to understand where I am making a mistake. Any comment is deeply appreciated. Well, if you are permuting Xvar and Yvar separately or sorting them (separately), then you cannot expect to get the same correlation again. Look at the formula and make an example for yourself with just, say,
Re: [R] search through a matrix
Thanks very much. I don't really understand the row() function. I looked in the reference but I don't really get it. It says: Description Returns a matrix of integers indicating their row number in a matrix-like object, or a factor indicating the row labels. Usage row(x, as.factor = FALSE) Arguments x a matrix-like object, that is one with a two-dimensional dim. I don't understand what row() does. And in the example in the documentation it says: x - matrix(1:12, 3, 4) # extract the diagonal of a matrix dx - x[row(x) == col(x)] dx [1] 1 5 9 I thought the single square bracket notation accepts a pair separated by a comma but I don't see how row(x)==col(x) produces that? Thanks again. onyourmark wrote: Hi. I have a 925 by 925 correlation matrix corM. I want to identify all variables that have correlation greater than 0.9. Can anyone suggest an R way of doing this? Thank you. -- View this message in context: http://www.nabble.com/search-through-a-matrix-tp23153538p23155104.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] search through a matrix
On Apr 21, 2009, at 8:33 AM, onyourmark wrote: Thanks very much. I don't really understand the row() function. I looked in the reference but I don't really get it. It says: Description Returns a matrix of integers indicating their row number in a matrix- like object, or a factor indicating the row labels. Usage row(x, as.factor = FALSE) Arguments x a matrix-like object, that is one with a two-dimensional dim. I don't understand what row() does. It returns a matrix of the same size as its arguments populated with the row numbers instead of the matrix elements. And in the example in the documentation it says: x - matrix(1:12, 3, 4) Now try row(x) # extract the diagonal of a matrix dx - x[row(x) == col(x)] dx [1] 1 5 9 I thought the single square bracket notation accepts a pair separated by a comma but I don't see how row(x)==col(x) produces that? The single square bracket notation can be used as [r,c] or without the comma which requires a logical index. In the second method the matrix entries get processed serially, column-wise. x - matrix(1:12, 3, 4) x[TRUE] [1] 1 2 3 4 5 6 7 8 9 10 11 12 row(x) [,1] [,2] [,3] [,4] [1,]1111 [2,]2222 [3,]3333 -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Help with R 2.9.0 and XML
After I installed R 2.9.0 and XML package, I get an error This application has failed to start because iconv.dll was not found. Re-installing the application may fix this problem. I already re-installed both but the problem persists. Does anyone know what is going on? Thanks. -- ___ Luís Orlindo Tedeschi mailto:luis.tedes...@gmail.com ___ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] search through a matrix
row() gives the row number of each element of the matrix, similarly col() gives the column number. By definition, diagonal elements are the elements correspond to same row and column numbers. That's how it works... x - matrix(1:12, 3, 4) x [,1] [,2] [,3] [,4] [1,] 12963 [2,] 11852 [3,] 10741 row(x) [,1] [,2] [,3] [,4] [1,]1111 [2,]2222 [3,]3333 col(x) [,1] [,2] [,3] [,4] [1,]1234 [2,]1234 [3,]1234 -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of onyourmark Sent: Tuesday, April 21, 2009 6:03 PM To: r-help@r-project.org Subject: Re: [R] search through a matrix Thanks very much. I don't really understand the row() function. I looked in the reference but I don't really get it. It says: Description Returns a matrix of integers indicating their row number in a matrix-like object, or a factor indicating the row labels. Usage row(x, as.factor = FALSE) Arguments x a matrix-like object, that is one with a two-dimensional dim. I don't understand what row() does. And in the example in the documentation it says: x - matrix(1:12, 3, 4) # extract the diagonal of a matrix dx - x[row(x) == col(x)] dx [1] 1 5 9 I thought the single square bracket notation accepts a pair separated by a comma but I don't see how row(x)==col(x) produces that? Thanks again. onyourmark wrote: Hi. I have a 925 by 925 correlation matrix corM. I want to identify all variables that have correlation greater than 0.9. Can anyone suggest an R way of doing this? Thank you. -- View this message in context: http://www.nabble.com/search-through-a-matrix-tp23153538p23155104.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. This e-mail may contain confidential and/or privileged i...{{dropped:10}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] search through a matrix
Hi again. Thanks. I get it now. So row(x) is a matrix with just the row number for each entry. How about x[row(x) == col(x)] ? Can the square bracket function take a matrix as its argument? If so, I guess I understand this statement. The argument is a boolean matrix. Thank you. David Winsemius wrote: On Apr 21, 2009, at 8:33 AM, onyourmark wrote: Thanks very much. I don't really understand the row() function. I looked in the reference but I don't really get it. It says: Description Returns a matrix of integers indicating their row number in a matrix- like object, or a factor indicating the row labels. Usage row(x, as.factor = FALSE) Arguments x a matrix-like object, that is one with a two-dimensional dim. I don't understand what row() does. It returns a matrix of the same size as its arguments populated with the row numbers instead of the matrix elements. And in the example in the documentation it says: x - matrix(1:12, 3, 4) Now try row(x) # extract the diagonal of a matrix dx - x[row(x) == col(x)] dx [1] 1 5 9 I thought the single square bracket notation accepts a pair separated by a comma but I don't see how row(x)==col(x) produces that? The single square bracket notation can be used as [r,c] or without the comma which requires a logical index. In the second method the matrix entries get processed serially, column-wise. x - matrix(1:12, 3, 4) x[TRUE] [1] 1 2 3 4 5 6 7 8 9 10 11 12 row(x) [,1] [,2] [,3] [,4] [1,]1111 [2,]2222 [3,]3333 -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/search-through-a-matrix-tp23153538p23155615.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
On Apr 21, 2009, at 5:55 AM, Duncan Murdoch wrote: On 21/04/2009 3:48 AM, Petr PIKAL wrote: Hi r-help-boun...@r-project.org napsal dne 20.04.2009 19:01:46: wolfgang.siewert wolfgang.siewert at gmail.com writes: There is a way around: round(0.7-0.3,1)==0.4 (TRUE) Obviously there is a problem with some combinations of decimal subtractions, that - we have the feeling - shouldt be solved. Oh no, not that one again! This was lecture two in my first computer course in 1968, but it seems to be gone the way of the dodo since than. Maybe that is because of Excel is so widespread now and gives expected results (it probably silently rounds all decimal numbers before calculation). I don't have Excel, but I expect OpenOffice duplicates its bugs pretty well. And in OpenOffice I see all sorts of bugs due to this, e.g. examples where x = y and y = z but x != z, cases where I can calculate a number like 1 + 4.e-15 and get something different from 1, but if I enter it directly as 1.004, it gets changed to 1. So it only gives expected results in some tests, not others. Duncan Murdoch As Dieter noted from our offlist exchange, this had been discussed previously back in 2003. Just to refresh memories: https://stat.ethz.ch/pipermail/r-help/2003-June/034565.html https://stat.ethz.ch/pipermail/r-help/2003-June/034860.html OO.org has replicated Excel's behavior to a fault. Thus: Spreadsheet Use - Brain to Porridge Just to update OO.org's behavior using version 3.0.1 on OSX: Formula: =4.145 * 100 + 0.5 Result: 415. Formula: =0.5 - 0.4 - 0.1 Result: 0. Formula: =(0.5 - 0.4 - 0.1) Result: 0. So nothing has changed in OO.org in five years. Somebody with Excel 2007 might want to try the 2nd and 3rd formula examples to see if using parens still makes a difference in the result as compared to the formula without the parens. FWIW, now that I am on OSX, I can add the following output using Numbers '09: Formula: =4.145 * 100 + 0.5 Result: 415. Formula: =0.5 - 0.4 - 0.1 Result: -2.77556E-17 Formula: =(0.5 - 0.4 - 0.1) Result: -2.77556E-17 It does look like R's behavior has changed since then. Using: R version 2.9.0 Patched (2009-04-18 r48348) on OSX: # This first example has changed. # Prior result was 414.94 print(4.145 * 100 + 0.5, digits = 20) [1] 415 formatC(4.145 * 100 + 0.5, format = E, digits = 20) [1] 4.14943157E+02 print(0.5 - 0.4 - 0.1, digits = 20) [1] -2.77555756156289e-17 formatC(0.5 - 0.4 - 0.1, format = E, digits = 20) [1] -2.77555756156289135106E-17 What is interesting is that: 4.145 * 100 + 0.5 == 415 [1] FALSE (4.145 * 100 + 0.5) - 415 [1] -5.684342e-14 all.equal(4.145 * 100 + 0.5, 415, 0) [1] Mean relative difference: 1.369721e-16 So it would appear that in the first R example above, the print() function has changed in a material fashion. HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] automatic exploration of all possible loglinear models?
Dieter Menne wrote: Christopher W. Ryan cryan at binghamton.edu writes: Is there a way to automate fitting and assessing loglinear models for several nominal variables . . . something akin to step or drop1 or add1 for linear or logistic regression? Not strictly for loglinear, but glm works with stepAIC. Make sure that in the field you are working this approach is an accepted ritual. Dieter Dieter was kind. I would say the following: Make sure your colleagues are interested in non-reproducible results, overfitting, incorrect P-values, and incorrect confidence limits before proceeding. Frank -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Matrix package,solve() errors and crashes
SS == Surendar Swaminathan surendar.swaminat...@gmail.com on Mon, 20 Apr 2009 12:10:47 -0700 writes: SS Hello All, I am working on graph object using IGRAPH SS package wanted to do Bonacich Power. This is my graph SS object. SS The file 'Graph.RData' (4.2 MB) is available for SS download at SS http://dropbox.unl.edu/uploads/20090424/cfe4fcb854bb17f2/Graph.RData SS Graph size SS Vertices: 20984 Edges: 326033 Directed: FALSE No graph SS attributes. Vertex attributes: name. No edge SS attributes. SS When I use bonacich power it goes out of memory SS Error in get.adjacency.dense(graph, type = type, attr = SS attr, names = names, : At vector.pmt:409 : cannot SS reserve space for vector, Out of memory how much memory (RAM) has your computer? I have tried in a few ways to reproduce your problem, but have not got any errors, neither when using the 'test.g' example you use below, nor by using bonpow.sparse(g) where 'g' comes from your 'Graph.RData' file (you mention above). But you have not provided (in your e-mail) a fully reproducible example, so I am not sure how exactly you got the memory problem (or even worse the crashes). Regards, Martin Maechler, ETH Zurich (co-maintainer of 'Matrix'). SS I got help from IGRAPH community to use sparse Matrix SS http://igraph.wikidot.com/r-recipes#toc6 SS bonpow.sparse - function(graph, nodes=V(graph), SS loops=FALSE, exponent=1, rescale=FALSE, tol=1e-07) { SS ## remove loops if requested if (!loops) { graph - SS simplify(graph, remove.multiple=FALSE, SS remove.loops=TRUE) } SS ## sparse adjacency matrix d - get.adjacency(graph, SS sparse=TRUE) SS ## sparse identity matrix id - SS spMatrix(vcount(graph), vcount(graph), SS i=1:vcount(graph), j=1:vcount(graph), x=rep(1, SS vcount(graph))) id - as(id, dgCMatrix) SS ## solve it ev - solve(id - exponent * d, tol=tol) SS %*% degree(graph, mode=out) SS if (rescale) { ev - ev/sum(ev) } else { ev - ev * SS sqrt(vcount(graph)/sum((ev)^2)) } SS ev[as.numeric(nodes) + 1] } SS ## test graph test.g - simplify(ba.game(1000,m=2)) SS ## test run system.time(bp1 - bonpow(test.g)) SS system.time(bp2 - bonpow.sparse(test.g)) SS ## check that they are the same max(abs(bp1-bp2)) I get SS following error and sometime it crashes. SS In solve(id - exponent * d, tol = tol) : Reached total SS allocation of 1535Mb: see help(memory.size).I increased SS the memory size and still it is not helpful SS Help on this would be great. SS These are steps I followed SS 1. Created graph object using Igraph version 0.6 on R SS 2.8.1 windows XP 2. Bonpow(g) 3. Bonpow.sparse function SS sessionInfo() SS R version 2.8.1 (2008-12-22) i386-pc-mingw32 locale: SS LC_COLLATE=English_United SS States.1252;LC_CTYPE=English_United SS States.1252;LC_MONETARY=English_United SS States.1252;LC_NUMERIC=C;LC_TIME=English_United SS States.1252 attached base packages: [1] stats graphics SS grDevices utils datasets methods base other attached SS packages: [1] Matrix_0.999375-23 lattice_0.17-20 SS igraph_0.6 loaded via a namespace (and not attached): SS [1] grid_2.8.1 SS Thanks in advance SS Nathan SS [[alternative HTML version deleted]] SS __ SS R-help@r-project.org mailing list SS https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do SS read the posting guide SS http://www.R-project.org/posting-guide.html and provide SS commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] search through a matrix
Just try to run x[row(x) == col(x)] . You'll see it's a vector. It's the same as diag(x) Dimitri On Tue, Apr 21, 2009 at 9:02 AM, onyourmark william...@gmail.com wrote: Hi again. Thanks. I get it now. So row(x) is a matrix with just the row number for each entry. How about x[row(x) == col(x)] ? Can the square bracket function take a matrix as its argument? If so, I guess I understand this statement. The argument is a boolean matrix. Thank you. David Winsemius wrote: On Apr 21, 2009, at 8:33 AM, onyourmark wrote: Thanks very much. I don't really understand the row() function. I looked in the reference but I don't really get it. It says: Description Returns a matrix of integers indicating their row number in a matrix- like object, or a factor indicating the row labels. Usage row(x, as.factor = FALSE) Arguments x a matrix-like object, that is one with a two-dimensional dim. I don't understand what row() does. It returns a matrix of the same size as its arguments populated with the row numbers instead of the matrix elements. And in the example in the documentation it says: x - matrix(1:12, 3, 4) Now try row(x) # extract the diagonal of a matrix dx - x[row(x) == col(x)] dx [1] 1 5 9 I thought the single square bracket notation accepts a pair separated by a comma but I don't see how row(x)==col(x) produces that? The single square bracket notation can be used as [r,c] or without the comma which requires a logical index. In the second method the matrix entries get processed serially, column-wise. x - matrix(1:12, 3, 4) x[TRUE] [1] 1 2 3 4 5 6 7 8 9 10 11 12 row(x) [,1] [,2] [,3] [,4] [1,] 1 1 1 1 [2,] 2 2 2 2 [3,] 3 3 3 3 -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- View this message in context: http://www.nabble.com/search-through-a-matrix-tp23153538p23155615.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] OT: Strange bootstrap approach to significance testing
Hi all, I'm reviewing a paper that employs a strange (to me) approach to a non-parametric significance testing. I'm familiar with permutation tests and bootstrapping confidence intervals around parameter estimates, but here they seem to be bootstrapping CIs for a manufactured Null F-value. They have a simple 3 groups between-Ss design and compute the F-value of the observed data. Then, they re-center each group's mean to zero then repeatedly: randomly resample values from each group (with replacement, doubling each group's size) and re-compute the F-value. The distribution of re-centered/resampled F-values is then used as a reference distribution within which the percentile of the observed F-value is computed. They determine that the observed F-value is outside the 95% confidence interval of the re-centered/resampled F-values and conclude a significant effect of group. In my head this makes some sense (though I'd think a straight permutation test would be a lot simpler), but having never heard of anything like this before I thought I'd see what others on this list think of the approach. Mike -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tr.im/mikes_public_calendar ~ Certainty is folly... I think. ~ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Package or packages for randomization in a clinical trial
On Apr 21, 2009, at 5:12 AM, John Sorkin wrote: Can anyone recommend a package that can be used to randomize subjects? I am looking for a generalized package, or several packages that can accomplish unrestricted randomization (i.e. simple random assignment) restricted randomization including stratified randomization, blocked randomization, and adaptive randomization. Thanks, John John, Take a look at the 'blockrand' and 'crossdes' packages on CRAN. If you need something beyond those, it is not overly difficult to write functions in R to do this. HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
try this 0.3-0.1 == 0.2 [1] FALSE round(0.3-0.1,1)==round(0.2,1) [1] TRUE liov 2009/4/21 Marc Schwartz marc_schwa...@me.com On Apr 21, 2009, at 5:55 AM, Duncan Murdoch wrote: On 21/04/2009 3:48 AM, Petr PIKAL wrote: Hi r-help-boun...@r-project.org napsal dne 20.04.2009 19:01:46: wolfgang.siewert wolfgang.siewert at gmail.com writes: There is a way around: round(0.7-0.3,1)==0.4 (TRUE) Obviously there is a problem with some combinations of decimal subtractions, that - we have the feeling - shouldt be solved. Oh no, not that one again! This was lecture two in my first computer course in 1968, but it seems to be gone the way of the dodo since than. Maybe that is because of Excel is so widespread now and gives expected results (it probably silently rounds all decimal numbers before calculation). I don't have Excel, but I expect OpenOffice duplicates its bugs pretty well. And in OpenOffice I see all sorts of bugs due to this, e.g. examples where x = y and y = z but x != z, cases where I can calculate a number like 1 + 4.e-15 and get something different from 1, but if I enter it directly as 1.004, it gets changed to 1. So it only gives expected results in some tests, not others. Duncan Murdoch As Dieter noted from our offlist exchange, this had been discussed previously back in 2003. Just to refresh memories: https://stat.ethz.ch/pipermail/r-help/2003-June/034565.html https://stat.ethz.ch/pipermail/r-help/2003-June/034860.html OO.org has replicated Excel's behavior to a fault. Thus: Spreadsheet Use - Brain to Porridge Just to update OO.org's behavior using version 3.0.1 on OSX: Formula: =4.145 * 100 + 0.5 Result: 415. Formula: =0.5 - 0.4 - 0.1 Result: 0. Formula: =(0.5 - 0.4 - 0.1) Result: 0. So nothing has changed in OO.org in five years. Somebody with Excel 2007 might want to try the 2nd and 3rd formula examples to see if using parens still makes a difference in the result as compared to the formula without the parens. FWIW, now that I am on OSX, I can add the following output using Numbers '09: Formula: =4.145 * 100 + 0.5 Result: 415. Formula: =0.5 - 0.4 - 0.1 Result: -2.77556E-17 Formula: =(0.5 - 0.4 - 0.1) Result: -2.77556E-17 It does look like R's behavior has changed since then. Using: R version 2.9.0 Patched (2009-04-18 r48348) on OSX: # This first example has changed. # Prior result was 414.94 print(4.145 * 100 + 0.5, digits = 20) [1] 415 formatC(4.145 * 100 + 0.5, format = E, digits = 20) [1] 4.14943157E+02 print(0.5 - 0.4 - 0.1, digits = 20) [1] -2.77555756156289e-17 formatC(0.5 - 0.4 - 0.1, format = E, digits = 20) [1] -2.77555756156289135106E-17 What is interesting is that: 4.145 * 100 + 0.5 == 415 [1] FALSE (4.145 * 100 + 0.5) - 415 [1] -5.684342e-14 all.equal(4.145 * 100 + 0.5, 415, 0) [1] Mean relative difference: 1.369721e-16 So it would appear that in the first R example above, the print() function has changed in a material fashion. HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Luis Iván Ortiz Valencia Estatístico Msc. ... Curriculum Lattes http://buscatextual.cnpq.br/buscatextual/visualizacv.jsp?id=K4778724J3 ... http://oplanetaliov.blogspot.com/ ... [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] search through a matrix
On Apr 21, 2009, at 9:02 AM, onyourmark wrote: Hi again. Thanks. I get it now. So row(x) is a matrix with just the row number for each entry. Yes. How about x[row(x) == col(x)] ? Can the square bracket function take a matrix as its argument? If so, I guess I understand this statement. The argument is a boolean matrix. It can and it often does. The argument to the [] operation can be either a logical vector or a numeric vector which gets treated as an index. Try x[8]. My understanding is that the matrix, row(x) == col(x), as an argument would first get coerced to a vector constructed from the column-wise projection of the matrix. You can read further on the help page in the section that discusses matrices and arrays: ?[ David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] ncdf,RNetCDF
Friends, Someone please share your experience in installing ncdf and RNetCDF in fedora10. I am not able make it work as it throws me the following error. My R version is 2.8.0. library(ncdf) unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Thanks, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] what is R best for? Evidence please.
Discussion about R capabilities seem to suggest that for optimization one may want to use some other software. As I've been working (with Ravi Varadhan mainly) to try to improve what is now called optim(), I needed to test some new methods, including a variant of CG. Coding only in R on a pretty vanilla 3Ghz PC, I was surprised that a generalized Rosenbrock function in n=5 parameters solved in less than 2 minutes. It seems a lot of my experiences in building routines that appear in optim() -- which were written on a machine with 8K (that's K) bytes for program AND data -- are really not valid any more. This has prompted us to try (and it not simple) to set up some infrastructure to get good (well, better?) measures of performance for optimization and nonlinear parameter estimation. Complications involve the variety of environments and configurations, as well as codings of test functions, choices of how to provide gradient information etc. Contact me off-line if you are interested in this, as we would like it to be relatively easy to use and share. That can come only by communication, and we want to have lots of real tests, and they take a fair bit of effort to set up in a standardized way. However, the main message here is to ask, as I was reasonably asked by another R worker, that people be much more cautious with conjectures. We can and should check our opinions by measuring, no matter what tasks or tools, when we give advice. JN __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Cross-Correlation function (CCF) issues
?? -- View this message in context: http://www.nabble.com/Cross-Correlation-function-%28CCF%29-issues-tp23145411p23156769.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] automatic exploration of all possible loglinear models?
Dieter Menne wrote: Christopher W. Ryan cryan at binghamton.edu writes: Is there a way to automate fitting and assessing loglinear models for several nominal variables . . . something akin to step or drop1 or add1 for linear or logistic regression? Not strictly for loglinear, but glm works with stepAIC. Make sure that in the field you are working this approach is an accepted ritual. Dieter There is also the package formerly known as dRedging: http://r-forge.r-project.org/projects/mumin/ -- View this message in context: http://www.nabble.com/automatic-exploration-of-all-possible--loglinear-models--tp23141046p23156882.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ncdf,RNetCDF
Hi, just quick googling gave me the hint that it can be ralated to SE Linux restrictions of Fedora linux. Just try to disable SE Linux feature. http://www.appistry.com/community/forums/content/cannot-restore-segment-prot-after-reloc-permission-denied Best regards, -- Andris subramanian bala.biophys...@gmail.com wrote: Friends, Someone please share your experience in installing ncdf and RNetCDF in fedora10. I am not able make it work as it throws me the following error. My R version is 2.8.0. library(ncdf) unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Thanks, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
A quarter-century ago I was part of the IEEE 854 group that tried to suggest a radix free arithmetic standard. One of the interests was decimal arithmetic. I actually have owned 2 decimal arithmetic computers, a North Star Horizon (with decimal hardware I had to build by hand from a kit) and a Radio Shack TRS80 Model 100, which used software BCD arithmetic. Interestingly, the same h/w for the TRS80 by NEC used binary FP. With the number of times the question of arithmetic coming up on this list, one wonders if a decimal arithmetic version of R, or other software too, might not be worthwhile. There are other issues that arise, of course, but binary floating point does upset folk used to working in base 10, and the I/O conversions do make things look messy. JN __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] broken example: lme() + multcomp() Tukey on repeated measures design
I am trying to do Tukey HSD comparisons on a repeated measures expt. I found the following example on r-help and quoted approvingly elsewhere. It is broken. Can anyone please tell me how to get it to work? I am using R 2.4.1. require(MASS) ## for oats data set require(nlme) ## for lme() require(multcomp) ## for multiple comparison stuff Aov.mod - aov(Y ~ N + V + Error(B/V), data = oats) Lme.mod - lme(Y ~ N + V, random = ~1 | B/V, data = oats) summary(Aov.mod) anova(Lme.mod) summary(Lme.mod) summary(glht(Lme.mod, linfct=mcp(V=Tukey))) Error in eval(expr, envir, enclos) : object Y not found Error in factor_contrasts(model) : no 'model.matrix' method for 'model' found! It all went ok until the last line, where the actual Tukey multiple comparisons was requested. Thanks very much for any help! Bill __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Bala subramanian wrote: Hi, I tried to install from four different mirrors. The result is the same. I dnt understand what the following error means. /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' I even tried to change the permission of ncdf.so but still i am not able to load the library. Nothing to do with the download. I think SElinux is getting in your way (so your unstated system could be RedHat Enterprise Linux?). Contact your sysadm or google for the error message. (One of the hits suggests that you might need to run chcon -t texrel_shlib_t ncdf.so but, well, no warranties...) Bala On Tue, Apr 21, 2009 at 2:03 PM, Dimitri Liakhovitski ld7...@gmail.comwrote: Try to install from scratch from a different mirror. Dimitri On Tue, Apr 21, 2009 at 6:21 AM, Bala subramanian bala.biophys...@gmail.com wrote: Hi, Yes ncdf needs netcdf, netcdf is already installed in my fedora10. I reinstalled ncdf with the following option install.packages('ncdf',dep=TRUE) library(ncdf) library(ncdf) Error in dyn.load(file, DLLpath = DLLpath, ...) : unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Again the same error, Bala On Tue, Apr 21, 2009 at 12:12 PM, Liviu Andronic landronim...@gmail.com wrote: On Tue, Apr 21, 2009 at 11:57 AM, Bala subramanian bala.biophys...@gmail.com wrote: Dear Simon, I installed the ncdf package in the way you suggested but still i got the same error i got before. I haves pasted below the installation log and errors. Does ncdf depend on netcdf [1]? If so, perhaps it is missing, or is not up-to-date. Also, it's a good habit to install.packages('ncdf', dep=TRUE). Regards, Liviu [1] http://www.unidata.ucar.edu/software/netcdf/ [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - (p.dalga...@biostat.ku.dk) FAX: (+45) 35327907 __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] broken example: lme() + multcomp() Tukey on repeated measures design
William Simpson william.a.simpson at gmail.com writes: I am trying to do Tukey HSD comparisons on a repeated measures expt. I found the following example on r-help and quoted approvingly elsewhere. It is broken. Can anyone please tell me how to get it to work? I am using R 2.4.1. That's the problem. Not the R-Version, but the likelihood that the multcomp package was not updated also, and lme definitively not in early version.s other attached packages: [1] multcomp_1.0-7 mvtnorm_0.9-5 nlme_3.1-90MASS_7.2-46 glht(Lme.mod, linfct=mcp(V=Tukey)) General Linear Hypotheses Multiple Comparisons of Means: Tukey Contrasts Linear Hypotheses: Estimate Marvellous - Golden.rain == 05.292 Victory - Golden.rain == 0 -6.875 Victory - Marvellous == 0 -12.167 Dieter __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] broken example: lme() + multcomp() Tukey on repeated measures design
On Apr 21, 2009, at 10:12 AM, William Simpson wrote: I am trying to do Tukey HSD comparisons on a repeated measures expt. I found the following example on r-help and quoted approvingly elsewhere. It is broken. Can anyone please tell me how to get it to work? I am using R 2.4.1. How many years have you had that version without updating? require(MASS) ## for oats data set require(nlme) ## for lme() require(multcomp) ## for multiple comparison stuff Aov.mod - aov(Y ~ N + V + Error(B/V), data = oats) Lme.mod - lme(Y ~ N + V, random = ~1 | B/V, data = oats) summary(Aov.mod) anova(Lme.mod) summary(Lme.mod) summary(glht(Lme.mod, linfct=mcp(V=Tukey))) Error in eval(expr, envir, enclos) : object Y not found Error in factor_contrasts(model) : no 'model.matrix' method for 'model' found! It is not broken when run in version 2.8.1. David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ncdf,RNetCDF
Hi Andris, THANK YOU SO MUCH. THIS WAS THE PROBLEM INDEED. When disable SE Linux, it worked fine. Thank you once again, Bala On Tue, Apr 21, 2009 at 4:01 PM, Andris Jankevics an...@osi.lv wrote: Hi, just quick googling gave me the hint that it can be ralated to SE Linux restrictions of Fedora linux. Just try to disable SE Linux feature. http://www.appistry.com/community/forums/content/cannot-restore-segment-prot-after-reloc-permission-denied Best regards, -- Andris subramanian bala.biophys...@gmail.com wrote: Friends, Someone please share your experience in installing ncdf and RNetCDF in fedora10. I am not able make it work as it throws me the following error. My R version is 2.8.0. library(ncdf) unable to load shared library '/usr/lib/R/library/ncdf/libs/ncdf.so': /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' Thanks, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Cross-Correlation function (CCF) issues
Are you trying to imply that people should be able to answer a question that included no data? As others have pointed out, our powers of telepathy are generally less than commonly assumed. On Apr 21, 2009, at 10:00 AM, manta wrote: ?? -- View this message in context: http://www.nabble.com/Cross-Correlation-function-%28CCF%29-issues-tp23145411p23156769.html Sent from the R help mailing list archive at Nabble.com. David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] ggplot2 - boxplot of variables / columns
Hi, ggplot/qplot is great - it has really helped me do some nice things. However, simple boxplot of different columns/variables is a bit tricky, because of (i think) qplot's generic Y conditional on X input form. Se below. # Some data: a - rnorm(100) b - rnorm(100,1,2) c - rnorm(100,2,0.5) # normal boxplot of a,b,c boxplot(a,b,c) # Looks good library(ggplot2) # loads qqplot2 # Tries do replicate the simple boxplot qplot(a,b,c, geom=boxplot) # Not good # Workaround d - c(a,b,c) e - c(rep(a,100),rep(b,100),rep(c,100)) qplot(e,d,geom=boxplot) # Works - but there must be a simpler way? What is the simple to compare multiple variables like this? thanks in advance __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
On Apr 21, 2009, at 9:28 AM, Peter Dalgaard wrote: Bala subramanian wrote: Hi, I tried to install from four different mirrors. The result is the same. I dnt understand what the following error means. /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' I even tried to change the permission of ncdf.so but still i am not able to load the library. Nothing to do with the download. I think SElinux is getting in your way (so your unstated system could be RedHat Enterprise Linux?). Contact your sysadm or google for the error message. (One of the hits suggests that you might need to run chcon -t texrel_shlib_t ncdf.so but, well, no warranties...) The OP did indicate that F10 was the platform: https://stat.ethz.ch/pipermail/r-help/attachments/20090421/65ee5d6e/attachment.pl so I would agree here that SELinux is almost certainly the culprit as it does control relocation. If you have the SE Troubleshoot daemon running, you should get a yellow star/badge symbol on one of your desktop dock panels. Typically, if you click on this, it will bring up a dialog window that will have some specific suggestions for resolving the problem. You can see some visual examples here: https://fedoraproject.org/wiki/Design/SETroubleshootUsabilityImprovements More info on SETroubleshoot here: https://fedorahosted.org/setroubleshoot/ HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Dear Marc and Peter, Thank you so much. In fact disabling the SElinux solved the problem. Bala On Tue, Apr 21, 2009 at 4:43 PM, Marc Schwartz marc_schwa...@me.com wrote: On Apr 21, 2009, at 9:28 AM, Peter Dalgaard wrote: Bala subramanian wrote: Hi, I tried to install from four different mirrors. The result is the same. I dnt understand what the following error means. /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' I even tried to change the permission of ncdf.so but still i am not able to load the library. Nothing to do with the download. I think SElinux is getting in your way (so your unstated system could be RedHat Enterprise Linux?). Contact your sysadm or google for the error message. (One of the hits suggests that you might need to run chcon -t texrel_shlib_t ncdf.so but, well, no warranties...) The OP did indicate that F10 was the platform: https://stat.ethz.ch/pipermail/r-help/attachments/20090421/65ee5d6e/attachment.pl so I would agree here that SELinux is almost certainly the culprit as it does control relocation. If you have the SE Troubleshoot daemon running, you should get a yellow star/badge symbol on one of your desktop dock panels. Typically, if you click on this, it will bring up a dialog window that will have some specific suggestions for resolving the problem. You can see some visual examples here: https://fedoraproject.org/wiki/Design/SETroubleshootUsabilityImprovements More info on SETroubleshoot here: https://fedorahosted.org/setroubleshoot/ HTH, Marc Schwartz [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Cross-Correlation function (CCF) issues
Sorry, my bad, i did not mean to 'be mean'. Here are the first five observations for three variables (dr1, dr2 and doil) dr1 1996-01-021996-01-031996-01-041996-01-051996-01-08 0.0005814396 -0.0023725000 -0.0072835915 0.0074536448 -0.0007004221 dr2 1996-01-031996-01-041996-01-051996-01-081996-01-09 -0.0029539396 -0.0049110915 0.0147372363 -0.0081540669 -0.0003020745 do1 1996-01-02 1996-01-03 1996-01-04 1996-01-05 1996-01-08 0.08 0.01 0.17 -0.03 0.00 As you can see, dr2 is nothing but the 1st difference of dr1. In my case, I'm trying to find out the cross-correlation between the two variables do1 and dr1 up to their 10th lag (i.e. do1 with do2, do3, ..., do10,dr1,dr2,...,dr10, and the same for dr1). Hope it helps, Marco David Winsemius wrote: Are you trying to imply that people should be able to answer a question that included no data? As others have pointed out, our powers of telepathy are generally less than commonly assumed. -- View this message in context: http://www.nabble.com/Cross-Correlation-function-%28CCF%29-issues-tp23145411p23157961.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] plotting with R
Friends, i) I am new to R. Kindly suggest some resources that has examples of plotting with R. ii) How to set number of tick marks and labels, i have a x axis ranging fro 1 to 21. By default R shows the tick marks at 5, 10,15,20. How can i change this. Thanks, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
David, Thanks for the suggestions. No, I did not label my dependent variable function. My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Bert: This is not homework. But I will remember to do my research before posting here. Aparna -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Monday, April 20, 2009 5:35 PM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 20, 2009, at 7:26 PM, Vemuri, Aparna wrote: I am not sure if this is an R-users question, but since most of you here are statisticians, I decided to give it a shot. You can omit the unnecessary preambles. I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.02950420.08429890.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] joined R-today
Bala, I would not advocate disabling SELinux entirely to resolve this issue. SELinux is one of the key security features of Fedora Linux and some other Linux distributions that have adopted it. It has already been demonstrated to be robust with respect to so-called zero-day attacks. I would first look to take a more surgical approach to resolve this particular issue with this package. Using the SETroubleshoot daemon is the easiest way to accomplish this as it will provide specific instructions. As with any security approach, there are trade-offs between security and usability, but SELinux as a mandatory access control system is well worthwhile keeping in place. BTW, as a future reference, there is a r-sig-fedora list: https://stat.ethz.ch/mailman/listinfo/r-sig-fedora HTH, Marc On Apr 21, 2009, at 9:56 AM, Bala subramanian wrote: Dear Marc and Peter, Thank you so much. In fact disabling the SElinux solved the problem. Bala On Tue, Apr 21, 2009 at 4:43 PM, Marc Schwartz marc_schwa...@me.com wrote: On Apr 21, 2009, at 9:28 AM, Peter Dalgaard wrote: Bala subramanian wrote: Hi, I tried to install from four different mirrors. The result is the same. I dnt understand what the following error means. /usr/lib/R/library/ncdf/libs/ncdf.so: cannot restore segment prot after reloc: Permission denied Error in library(ncdf) : .First.lib failed for 'ncdf' I even tried to change the permission of ncdf.so but still i am not able to load the library. Nothing to do with the download. I think SElinux is getting in your way (so your unstated system could be RedHat Enterprise Linux?). Contact your sysadm or google for the error message. (One of the hits suggests that you might need to run chcon -t texrel_shlib_t ncdf.so but, well, no warranties...) The OP did indicate that F10 was the platform: https://stat.ethz.ch/pipermail/r-help/attachments/20090421/65ee5d6e/attachment.pl so I would agree here that SELinux is almost certainly the culprit as it does control relocation. If you have the SE Troubleshoot daemon running, you should get a yellow star/badge symbol on one of your desktop dock panels. Typically, if you click on this, it will bring up a dialog window that will have some specific suggestions for resolving the problem. You can see some visual examples here: https://fedoraproject.org/wiki/Design/SETroubleshootUsabilityImprovements More info on SETroubleshoot here: https://fedorahosted.org/setroubleshoot/ HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ggplot2 - boxplot of variables / columns
On Apr 21, 2009, at 10:42 AM, Andreas Christoffersen wrote: Hi, ggplot/qplot is great - it has really helped me do some nice things. However, simple boxplot of different columns/variables is a bit tricky, because of (i think) qplot's generic Y conditional on X input form. Se below. # Some data: a - rnorm(100) b - rnorm(100,1,2) c - rnorm(100,2,0.5) # normal boxplot of a,b,c boxplot(a,b,c) # Looks good library(ggplot2) # loads qqplot2 # Tries do replicate the simple boxplot qplot(a,b,c, geom=boxplot) # Not good # Workaround d - c(a,b,c) e - c(rep(a,100),rep(b,100),rep(c,100)) qplot(e,d,geom=boxplot) # Works - but there must be a simpler way? qplot(ind, values, data=stack(data.frame(a,b,c)), geom=boxplot) I first tried stack(list(a,b,c)) but did not get the expected results. If anyone wants to enlighten me on why, I would be happy to offer a rewrite of the stack help page that clarifies my inability to parse it correctly in its current incarnation. What is the simple to compare multiple variables like this? David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] ggplot2 - boxplot of variables / columns
Dear Andreas, melt() and cast() are nice tools for this kind of problems. They both reside in the reshape package that automatic loaded when ggplot2 is. a - rnorm(100) b - rnorm(100,1,2) c - rnorm(100,2,0.5) ds - data.frame(a = a, b = b, c = c) library(ggplot2) # loads qqplot2 ggplot(melt(ds), aes(x = variable, y = value)) + geom_boxplot() HTH, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 thierry.onkel...@inbo.be www.inbo.be To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of. ~ Sir Ronald Aylmer Fisher The plural of anecdote is not data. ~ Roger Brinner The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. ~ John Tukey -Oorspronkelijk bericht- Van: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] Namens Andreas Christoffersen Verzonden: dinsdag 21 april 2009 16:42 Aan: r-help@r-project.org Onderwerp: [R] ggplot2 - boxplot of variables / columns Hi, ggplot/qplot is great - it has really helped me do some nice things. However, simple boxplot of different columns/variables is a bit tricky, because of (i think) qplot's generic Y conditional on X input form. Se below. # Some data: a - rnorm(100) b - rnorm(100,1,2) c - rnorm(100,2,0.5) # normal boxplot of a,b,c boxplot(a,b,c) # Looks good library(ggplot2) # loads qqplot2 # Tries do replicate the simple boxplot qplot(a,b,c, geom=boxplot) # Not good # Workaround d - c(a,b,c) e - c(rep(a,100),rep(b,100),rep(c,100)) qplot(e,d,geom=boxplot) # Works - but there must be a simpler way? What is the simple to compare multiple variables like this? thanks in advance __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document. The views expressed in this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Cross-Correlation function (CCF) issues
We still have an inadequate characterization of the data to answe the question ( as I remember it from yesterday). Missing, for example, is any information about lengths which would seem essential since (as I remember) you wantted to know why the result was so short. Why not put in a full working example with an extract of the data. Suggest you try using dput as a method of creating a working example. That way we (and the R interpreter) would get labels and class information. -- David Winsemius On Apr 21, 2009, at 10:56 AM, manta wrote: Sorry, my bad, i did not mean to 'be mean'. Here are the first five observations for three variables (dr1, dr2 and doil) dr1 1996-01-021996-01-031996-01-041996-01-051996-01-08 0.0005814396 -0.0023725000 -0.0072835915 0.0074536448 -0.0007004221 dr2 1996-01-031996-01-041996-01-051996-01-081996-01-09 -0.0029539396 -0.0049110915 0.0147372363 -0.0081540669 -0.0003020745 do1 1996-01-02 1996-01-03 1996-01-04 1996-01-05 1996-01-08 0.08 0.01 0.17 -0.03 0.00 As you can see, dr2 is nothing but the 1st difference of dr1. In my case, I'm trying to find out the cross-correlation between the two variables do1 and dr1 up to their 10th lag (i.e. do1 with do2, do3, ..., do10,dr1,dr2,...,dr10, and the same for dr1). Hope it helps, Marco David Winsemius wrote: Are you trying to imply that people should be able to answer a question that included no data? As others have pointed out, our powers of telepathy are generally less than commonly assumed. -- View this message in context: http://www.nabble.com/Cross-Correlation-function-%28CCF%29-issues-tp23145411p23157961.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] plotting with R
On Apr 21, 2009, at 11:07 AM, Bala subramanian wrote: Friends, i) I am new to R. Kindly suggest some resources that has examples of plotting with R. http://addictedtor.free.fr/graphiques/ http://lmdvr.r-forge.r-project.org/figures/figures.html ii) How to set number of tick marks and labels, i have a x axis ranging fro 1 to 21. By default R shows the tick marks at 5, 10,15,20. How can i change this. ?axis -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
Aparna, I should have been more explicit. Run ?lm . You'll see this: lm(formula, data, subset, weights, na.action, method = qr, model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, contrasts = NULL, offset, ...) So, in addition to specifying the formula, you have to specify the data frame in which you keep your variables. I assume they are in a data frame? (unless for some reasons you keep all variables as separate vectors). So, after you wrote the formula, you have to indicate the name of the data frame, for example MyData: model1-lm(PBW~SO4+NO3+NH4, MyData) Dimitri On Tue, Apr 21, 2009 at 11:12 AM, Vemuri, Aparna avem...@epri.com wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Bert: This is not homework. But I will remember to do my research before posting here. Aparna -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Monday, April 20, 2009 5:35 PM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 20, 2009, at 7:26 PM, Vemuri, Aparna wrote: I am not sure if this is an R-users question, but since most of you here are statisticians, I decided to give it a shot. You can omit the unnecessary preambles. I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
The variables are all in separate vectors. -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 8:26 AM To: Vemuri, Aparna Cc: David Winsemius; r-help@r-project.org Subject: Re: [R] Fitting linear models Aparna, I should have been more explicit. Run ?lm . You'll see this: lm(formula, data, subset, weights, na.action, method = qr, model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, contrasts = NULL, offset, ...) So, in addition to specifying the formula, you have to specify the data frame in which you keep your variables. I assume they are in a data frame? (unless for some reasons you keep all variables as separate vectors). So, after you wrote the formula, you have to indicate the name of the data frame, for example MyData: model1-lm(PBW~SO4+NO3+NH4, MyData) Dimitri On Tue, Apr 21, 2009 at 11:12 AM, Vemuri, Aparna avem...@epri.com wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Bert: This is not homework. But I will remember to do my research before posting here. Aparna -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Monday, April 20, 2009 5:35 PM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 20, 2009, at 7:26 PM, Vemuri, Aparna wrote: I am not sure if this is an R-users question, but since most of you here are statisticians, I decided to give it a shot. You can omit the unnecessary preambles. I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] plotting with R
Lookup the R graph gallery. Also, any R tutorial should have some rudimentary plots. Type ?plot Into the prompt to get help with the command. -stephen Sent via BlackBerry from T-Mobile -Original Message- From: Bala subramanian bala.biophys...@gmail.com Date: Tue, 21 Apr 2009 17:07:24 To: r-help@r-project.org Subject: [R] plotting with R Friends, i) I am new to R. Kindly suggest some resources that has examples of plotting with R. ii) How to set number of tick marks and labels, i have a x axis ranging fro 1 to 21. By default R shows the tick marks at 5, 10,15,20. How can i change this. Thanks, Bala [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
Are they of the same length? On Tue, Apr 21, 2009 at 11:31 AM, Vemuri, Aparna avem...@epri.com wrote: The variables are all in separate vectors. -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 8:26 AM To: Vemuri, Aparna Cc: David Winsemius; r-help@r-project.org Subject: Re: [R] Fitting linear models Aparna, I should have been more explicit. Run ?lm . You'll see this: lm(formula, data, subset, weights, na.action, method = qr, model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, contrasts = NULL, offset, ...) So, in addition to specifying the formula, you have to specify the data frame in which you keep your variables. I assume they are in a data frame? (unless for some reasons you keep all variables as separate vectors). So, after you wrote the formula, you have to indicate the name of the data frame, for example MyData: model1-lm(PBW~SO4+NO3+NH4, MyData) Dimitri On Tue, Apr 21, 2009 at 11:12 AM, Vemuri, Aparna avem...@epri.com wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Bert: This is not homework. But I will remember to do my research before posting here. Aparna -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Monday, April 20, 2009 5:35 PM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 20, 2009, at 7:26 PM, Vemuri, Aparna wrote: I am not sure if this is an R-users question, but since most of you here are statisticians, I decided to give it a shot. You can omit the unnecessary preambles. I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
Yes, they are all of the same length. -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 8:32 AM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models Are they of the same length? On Tue, Apr 21, 2009 at 11:31 AM, Vemuri, Aparna avem...@epri.com wrote: The variables are all in separate vectors. -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 8:26 AM To: Vemuri, Aparna Cc: David Winsemius; r-help@r-project.org Subject: Re: [R] Fitting linear models Aparna, I should have been more explicit. Run ?lm . You'll see this: lm(formula, data, subset, weights, na.action, method = qr, model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, contrasts = NULL, offset, ...) So, in addition to specifying the formula, you have to specify the data frame in which you keep your variables. I assume they are in a data frame? (unless for some reasons you keep all variables as separate vectors). So, after you wrote the formula, you have to indicate the name of the data frame, for example MyData: model1-lm(PBW~SO4+NO3+NH4, MyData) Dimitri On Tue, Apr 21, 2009 at 11:12 AM, Vemuri, Aparna avem...@epri.com wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Bert: This is not homework. But I will remember to do my research before posting here. Aparna -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Monday, April 20, 2009 5:35 PM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 20, 2009, at 7:26 PM, Vemuri, Aparna wrote: I am not sure if this is an R-users question, but since most of you here are statisticians, I decided to give it a shot. You can omit the unnecessary preambles. I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
On Apr 21, 2009, at 11:12 AM, Vemuri, Aparna wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. That was from my error in reading your call to lm. In my defense I am reasonably sure the proper assignment to arguments is lm(formula= ...) rather than lm(function= ...). My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Did you get the expected results with; model1-lm(formula=PBW~SO4+NO3+NH4+0) You could, of course, provide either the data or the results of str() applied to each of the variables and then we could all stop guessing. Aparna I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.02950420.08429890.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] My surprising experience in trying out REvolution's R
I care a lot about R's speed. So I decided to give REvolution's R (http://revolution-computing.com/) a try, which bills itself as an optimized R. Note that I used the free version. My machine is a Intel core 2 duo under Windows XP professional. The code I run is in the end of this post. First, the regular R 1.9. It takes 2 minutes and 6 seconds, CPU usage 50% Next, REvolution's R. It takes 2 minutes and 10 seconds, CPU usage 100%. In other words, REvolution's R consumes double the CPU with slightly less speed. The above has been replicated a few times (as a statistician of course). Anyone has any insight on this? Anyway, my high hope was dashed. rm(list=ls(all=TRUE)) library(MASS); ###small and common functions## glmm.sample.y - function(b, D.u, x, z) { pp - matrix(0, nrow = n, ncol = m); zero - numeric(n.random); ran - t( mvrnorm(m, zero, D.u) ); for(j in 1:m) pp[, j] - as.vector( x[, , j] %*% b + z[, , j] %*% ran[ ,j] ); pp - exp(pp)/( 1+exp(pp) ); y - runif(m*n); ifelse(ypp, 1, 0); } # quadratic.form - function(A, x) { return( as.vector(x %*% A %*% x) ) } quadratic.form2 - function(A, x) { x - chol(A) %*% x; colSums(x*x) } ### REML.b.D.u - function(b, D.u, x, y, z, n.iter) { u.mean.initial - array( 0, c(n.random, m) ); TT - matrix(0, n.random, n.random); for(i in 1:n.iter) { TT - TT*(1-1/i) + bias(b, D.u, x, z)/i; obj - sample.u(b, D.u, x, y, z, u.mean.initial); b - b - solve(obj$Hessian, obj$score); D.u - obj$uu + TT; u.mean.initial - obj$u.mean.initial; print(i); print(date()); print(inside REML); print(TT); if(i==n.iter) write(TT, file=c:/liao/reml/simu50_8.dat, append=T); print(D.u); } list(b=b, D.u=D.u); } bias - function(b, D.u, x, z) { yy - glmm.sample.y(b, D.u, x, z); #this is the sampling stage obj1 - mle(b, D.u, x, yy, z, F, F, n.iter=50); obj2 - mle(b, D.u, x, yy, z, T, F, n.iter=50); obj1$uu - obj2$uu } ## mle - function(b, D.u, x, y, z, indx1, indx2, n.iter) { u.mean.initial - array( 0, c(n.random, m) ); for(i in 1:n.iter) { obj - sample.u(b, D.u, x, y, z, u.mean.initial); if(indx1) b - b - solve(obj$Hessian, obj$score); if(indx2) D.u - obj$uu; u.mean.initial - obj$u.mean.initial; } list(b=b, D.u=D.u, uu=obj$uu, u.mean.initial=u.mean.initial); } ## sample.u - function(b, D.u, x, y, z, u.mean.initial) { D.u.inv - solve(D.u); uu - matrix(0, n.random, n.random); score - numeric(n.fixed); Hessian - matrix(0, n.fixed, n.fixed); for(i in 1:m) { ada.part - as.vector(x[, , i] %*% b); obj - sample.u.one( D.u.inv, ada.part, x[, , i], y[, i], z[, ,i], u.mean.initial[, i] ); uu - uu + obj$uu; score - score + obj$score; Hessian - Hessian + obj$Hessian; u.mean.initial[, i] - obj$u.mean.initial; } uu - uu/m; list(uu=uu, score=score, Hessian=Hessian, u.mean.initial=u.mean.initial); } ## sample.u.one - function(D.u.inv, ada.part, x, y, z, u.mean.initial) #ada.part, x, y, z for one subject only { fn - log.like(y, z, D.u.inv, ada.part); obj - optim(u.mean.initial, fn, control=list(fnscale=-1), hessian = T) u.mean.initial - obj$par; var.root - solve(-obj$hessian); var.root - t( chol(var.root) ); u - matrix( rnorm(n.random*n.simu), nrow=n.random, ncol=n.simu ); log.prob1 - -colSums(u*u)/2; u - u.mean.initial + var.root %*% u; ada - exp( ada.part + z %*% u ); ada - ada/(1+ada); #it is probability now log.prob2 - colSums( y*log(ada) + (1-y)*log(1-ada) ); log.prob2 - -quadratic.form2(D.u.inv, u)/2 + log.prob2; weight - exp(log.prob2 - log.prob1); weight - weight/sum(weight); ada - t(ada); pi - colSums(ada*weight); product - colSums( ada*(1-ada)*weight ); score - as.vector( (y - pi) %*% x ); Hessian - -t(x) %*% ( rep(product, n.fixed)*x ); obj - cov.wt( t(u), weight, center=T ); uu - obj$cov + obj$center %*% t(obj$center); list(uu=uu, score=score, Hessian=Hessian, u.mean.initial=obj$center); } # log.like - function(y, z, D.u.inv, ada.part) { function(u) { ada - exp( ada.part + z %*% u ); ada - ada/(1+ada); sum( y*log(ada) + (1-y)*log(1-ada) ) -quadratic.form(D.u.inv, u)/2;
Re: [R] ggplot2 - boxplot of variables / columns
David, you solution qplot(ind, values, data=stack(data.frame(a,b,c)), geom=boxplot) Works a treat - thank you! Thierry, your solution ds - data.frame(a = a, b = b, c = c) library(ggplot2) # loads qqplot2 ggplot(melt(ds), aes(x = variable, y = value)) + geom_boxplot() Also works. I can even combine the two solutions, using stack / melt interchangeably. I am very glad that today I both learned melt/stack and Thierry has made me curious as to the finer ggplots fine tuning wizardry in ggplot vs. the simple qplot. Again - thx. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
Attached are the first hundred rows of my data in comma separated format. Forcing the regression line through the origin, still does not give a coefficient on the last independent variable. Also, I don't mind if there is a coefficient on the dependent axis. I just want all of the variables to have coefficients in the regression equation or a at least a consistent result, irrespective of the order of input information. -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Tuesday, April 21, 2009 8:38 AM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 21, 2009, at 11:12 AM, Vemuri, Aparna wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. That was from my error in reading your call to lm. In my defense I am reasonably sure the proper assignment to arguments is lm(formula= ...) rather than lm(function= ...). My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Did you get the expected results with; model1-lm(formula=PBW~SO4+NO3+NH4+0) You could, of course, provide either the data or the results of str() applied to each of the variables and then we could all stop guessing. Aparna I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.02950420.08429890.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT PBW,SO4,NO3,NH4 0.063,3.218,0.089,1.233 0.036,1.837,0.088,0.714 0.032,1.655,0.188,0.675 0.024,1.249,0.132,0.507 0.053,2.72,0.125,1.056 0.055,2.775,0.028,1.049 0.06,3.035,0.022,1.144 0.029,1.491,0.03,0.568 0.038,1.911,0.049,0.731 0.033,1.675,0.042,0.64 0.043,2.183,0.02,0.825 0.024,1.228,0.03,0.469 0.042,2.159,0.055,0.825 0.04,2.047,0.036,0.778 0.008,0.413,0.027,0.163 0.152,7.678,0.04,2.891 0.106,5.383,0.019,2.024 0.076,3.842,0.071,1.461 0.029,1.474,0.01,0.556 0.048,2.454,0.028,0.928 0.064,3.239,0.03,1.223 0.074,3.761,0.054,1.426 0.074,3.731,0.062,1.417 0.216,10.953,0.027,4.115 0.158,8.095,0.384,3.147 0.065,3.441,0.614,1.468 0.027,1.383,0.079,0.541 0.084,4.242,0.072,1.612 0.096,4.875,0.049,1.842 0.052,2.617,0.052,0.996 0.025,1.257,0.033,0.481 0.057,2.913,0.042,1.104 0.081,4.097,0.051,1.551 0.028,1.431,0.009,0.539 0.134,6.799,0.05,2.564 0.095,4.808,0.056,1.819 0.023,1.173,0.025,0.447 0.024,1.23,0.045,0.474 0.021,1.123,0.148,0.464 0.027,1.366,0.073,0.534 0.049,2.467,0.045,0.938 0.05,2.537,0.061,0.969 0.06,3.033,0.077,1.16 0.093,4.708,0.106,1.796 0.095,4.838,0.073,1.835 0.086,4.342,0.07,1.649 0.032,1.725,0.465,0.782 0.041,2.304,0.881,1.12 0.106,5.919,1.955,2.787 0.062,3.146,0.059,1.197 0.104,5.285,0.065,2.001 0.017,0.849,0.009,0.321 0.03,1.514,0.031,0.577 0.032,1.651,0.058,0.636 0.039,1.975,0.021,0.747 0.048,2.434,0.045,0.926 0.045,2.283,0.073,0.877 0.065,3.294,0.009,1.238 0.105,5.323,0.028,2.004 0.029,1.45,0.02,0.55 0.062,3.202,0.156,1.246 0.119,6.018,0.009,2.259 0.147,7.418,0.009,2.785 0.042,2.116,0.061,0.811 0.03,1.589,0.218,0.659 0.043,2.177,0.055,0.832 0.067,3.377,0.009,1.269 0.051,2.564,0.009,0.964 0.063,3.194,0.041,1.21 0.082,4.183,0.035,1.578 0.047,2.362,0.026,0.893 0.063,3.204,0.089,1.227 0.062,3.151,0.049,1.196
[R] Blanchard Quah/Impulse Response Functions
Dear all, How can I compute an impulse response function for a BQ (or any SVAR) in which the endogenous variable for output is given in log-difference, and I want the response in terms of the level of output, as done in Blanchard and Quah(1989)? Does the function irf require that the response is exactly as given in the set of endogenous variables? Thank you, Laura Carvalho __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
I am not sure what the problem is. I found no errors: data-read.csv(file.choose()) # I had to change your file extension to .csv first dim(data) names(data) lapply(data,function(x){sum(is.na(x))}) lm.model.1-lm(PBW~SO4+NO3+NH4,data) lm.model.2-lm(PBW~SO4+NH4+NO3,data) print(lm.model.1) # Getting nice results print(lm.model.2) # Getting same results # Another method (gets exactly the same results): library(Design) ols.model.1-ols(PBW~SO4+NO3+NH4,data) ols.model.2-ols(PBW~SO4+NH4+NO3,data) Dimitri On Tue, Apr 21, 2009 at 11:50 AM, Vemuri, Aparna avem...@epri.com wrote: Attached are the first hundred rows of my data in comma separated format. Forcing the regression line through the origin, still does not give a coefficient on the last independent variable. Also, I don't mind if there is a coefficient on the dependent axis. I just want all of the variables to have coefficients in the regression equation or a at least a consistent result, irrespective of the order of input information. -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Tuesday, April 21, 2009 8:38 AM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 21, 2009, at 11:12 AM, Vemuri, Aparna wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. That was from my error in reading your call to lm. In my defense I am reasonably sure the proper assignment to arguments is lm(formula= ...) rather than lm(function= ...). My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Did you get the expected results with; model1-lm(formula=PBW~SO4+NO3+NH4+0) You could, of course, provide either the data or the results of str() applied to each of the variables and then we could all stop guessing. Aparna I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Polyspline Integration
Hello, I wrote the function below to integrate polysplines and thought that it may be useful to others. Please consider this code released under the GPL2 or later. Thanks, Bill integrate.polySpline.R Notice: This e-mail message, together with any attachments, contains information of Merck Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp Dohme or MSD and in Japan, as Banyu - direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
On Apr 21, 2009, at 10:37 AM, David Winsemius wrote: On Apr 21, 2009, at 11:12 AM, Vemuri, Aparna wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. That was from my error in reading your call to lm. In my defense I am reasonably sure the proper assignment to arguments is lm(formula= ...) rather than lm(function= ...). My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Did you get the expected results with; model1-lm(formula=PBW~SO4+NO3+NH4+0) You could, of course, provide either the data or the results of str() applied to each of the variables and then we could all stop guessing. I am going to take a wild stab in the dark here and suggest that 'NH4' is exactly correlated to or even identical to one of the other IVs used in the formula. set.seed(1) PBW - rnorm(100) SO4 - rnorm(100) NO3 - rnorm(100) NH4 - rnorm(100) lm(PBW ~ SO4 + NO3 + NH4) Call: lm(formula = PBW ~ SO4 + NO3 + NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.11065 -0.00273 0.02096 -0.04826 Now watch: NH4 - NO3 * 1.5 lm(PBW ~ SO4 + NO3 + NH4) Call: lm(formula = PBW ~ SO4 + NO3 + NH4) Coefficients: (Intercept) SO4 NO3 NH4 1.084e-01 -7.871e-051.596e-02 NA cor(cbind(SO4, NO3, NH4)) SO4 NO3 NH4 SO4 1. -0.04953621 -0.04953621 NO3 -0.04953621 1. 1. NH4 -0.04953621 1. 1. I suspect that there is a collinearity problem here. Aparna, post back with the correlation matrix of your IV's (full data set) and that should either support or refute my theory. If supported and you use: summary(lm(PBW ~ SO4 + NO3 + NH4)) Call: lm(formula = PBW ~ SO4 + NO3 + NH4) Residuals: Min 1Q Median 3Q Max -2.30129 -0.60350 0.01765 0.58513 2.27806 Coefficients: (1 not defined because of singularities) Estimate Std. Error t value Pr(|t|) (Intercept) 1.084e-01 9.083e-02 1.1940.236 SO4 -7.871e-05 9.531e-02 -0.0010.999 NO3 1.596e-02 8.827e-02 0.1810.857 NH4 NA NA NA NA Residual standard error: 0.9073 on 97 degrees of freedom Multiple R-squared: 0.0003379, Adjusted R-squared: -0.02027 F-statistic: 0.01639 on 2 and 97 DF, p-value: 0.9837 Note the warning message about singularities for NH4. BTW, as an aside, picking variables for a model based upon their correlation with the DV is not a good way to go. You might want to pick up a copy of Frank's book Regression Modeling Strategies: http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/RmS HTH, Marc Schwartz __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
Can we see your data to be able to replicate the error? Or maybe a subset of data with some fake variable names? On Tue, Apr 21, 2009 at 11:32 AM, Vemuri, Aparna avem...@epri.com wrote: Yes, they are all of the same length. -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 8:32 AM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models Are they of the same length? On Tue, Apr 21, 2009 at 11:31 AM, Vemuri, Aparna avem...@epri.com wrote: The variables are all in separate vectors. -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 8:26 AM To: Vemuri, Aparna Cc: David Winsemius; r-help@r-project.org Subject: Re: [R] Fitting linear models Aparna, I should have been more explicit. Run ?lm . You'll see this: lm(formula, data, subset, weights, na.action, method = qr, model = TRUE, x = FALSE, y = FALSE, qr = TRUE, singular.ok = TRUE, contrasts = NULL, offset, ...) So, in addition to specifying the formula, you have to specify the data frame in which you keep your variables. I assume they are in a data frame? (unless for some reasons you keep all variables as separate vectors). So, after you wrote the formula, you have to indicate the name of the data frame, for example MyData: model1-lm(PBW~SO4+NO3+NH4, MyData) Dimitri On Tue, Apr 21, 2009 at 11:12 AM, Vemuri, Aparna avem...@epri.com wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Bert: This is not homework. But I will remember to do my research before posting here. Aparna -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Monday, April 20, 2009 5:35 PM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 20, 2009, at 7:26 PM, Vemuri, Aparna wrote: I am not sure if this is an R-users question, but since most of you here are statisticians, I decided to give it a shot. You can omit the unnecessary preambles. I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] DNAcopy package in R
Hi All: Has anyone analyzed Illumina 550K data using DNAcopy package? I have around 2300 samples. According to Venkatraman and Olshen 2007 paper, it needs about 25 min to run a single sample for Affymetrix 100K. I am afraid it will take too long to analyzed my data. Anybody has an idea? Thanks! Best, -- View this message in context: http://www.nabble.com/DNAcopy-package-in-R-tp23158907p23158907.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] fitting assimptotic decaing with - and + on X
Dear all, I need to adjust a decaing function between preference as function of distance (dst). As you can see below, the X values range from highly negative to higly positive. The preferences range from 1 to ~0. I tryed to use drc() packages, but I I am not wrong, the packages is developed mainly to responde-dose, so the X is from 0-infinite. I need to fit the data and get an assintoptic (inverse-S) shape. Any help are welcome. milton brazil-toronto --- sp.coredep.attr-read.table(stdin(), head=T, sep=,) species,dst,preference,probdye coredep,-7500,1,0.001 coredep,-1000,0.95,0.002 coredep,-500,0.9,0.003 coredep,-250,0.8,0.005 coredep,-120,0.7,0.01 coredep,-100,0.5,0.02 coredep,-90,0.3,0.03 coredep,-60,0.2,0.05 coredep,-30,0.15,0.065 coredep,0,0.1,0.08 coredep,30,0.09,0.1 coredep,60,0.08,0.15 coredep,90,0.07,0.2 coredep,120,0.06,0.3 coredep,240,0.05,0.4 coredep,500,0.04,0.5 coredep,1000,0.02,0.6 coredep,7500,0.01,0.7 plot(preference~as.factor(dst), data=sp.coredep.attr, type=n, cex=0.001, col=transparent, xlab=Distance (m), main=CORE Dependent) lines(sp.coredep.attr$preference, type=b, lwd=2, col=blue) abline(v=10, col=blue, lty=3) text(10, 1, EDGE, cex=1.5) text(7, 1, CORE, cex=1.5, col=green) text(13, 1, MATRIX, cex=1.5, col=red) text(4,0.65, Preference, col=blue) text(16,0.65, Prob.Die, col=red) lines(sp.coredep.attr$probdye, type=b, lwd=2, col=red, lty=2) [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] bug when subtracting decimals?
Have you read the posting guide and the FAQs? If you do not get a reply within two days, you may want to look at both and think about reformulating your query. Oh, and while you are at it, look through the archives, a lot of questions have already been asked and answered before. As I say every time someone brings this up, there are currently ~130 printed pages of FAQs. Reading all that seems a rather large burden on the novice poster. Hadley -- http://had.co.nz/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Fitting linear models
Thanks Dimitri! Following exactly what you did, I wrote all my individual variable vectors to a data frame and used lm(formula,data) and this time it works for me too. Marc, your theory is correct.NH4 variable shares a strong correlation with one of the IV along with the DV. SO4 NO3 NH4 PBW SO4 1 -0.0867 0.999 0.999 NO3 -0.0867 1 -0.0527 -0.0938 NH4 0.999 -0.0527 1 0.999 PBW 0.999 -0.0938 0.999 1 Aparna -Original Message- From: Dimitri Liakhovitski [mailto:ld7...@gmail.com] Sent: Tuesday, April 21, 2009 9:02 AM To: Vemuri, Aparna Cc: r-help@r-project.org; David Winsemius Subject: Re: [R] Fitting linear models I am not sure what the problem is. I found no errors: data-read.csv(file.choose()) # I had to change your file extension to .csv first dim(data) names(data) lapply(data,function(x){sum(is.na(x))}) lm.model.1-lm(PBW~SO4+NO3+NH4,data) lm.model.2-lm(PBW~SO4+NH4+NO3,data) print(lm.model.1) # Getting nice results print(lm.model.2) # Getting same results # Another method (gets exactly the same results): library(Design) ols.model.1-ols(PBW~SO4+NO3+NH4,data) ols.model.2-ols(PBW~SO4+NH4+NO3,data) Dimitri On Tue, Apr 21, 2009 at 11:50 AM, Vemuri, Aparna avem...@epri.com wrote: Attached are the first hundred rows of my data in comma separated format. Forcing the regression line through the origin, still does not give a coefficient on the last independent variable. Also, I don't mind if there is a coefficient on the dependent axis. I just want all of the variables to have coefficients in the regression equation or a at least a consistent result, irrespective of the order of input information. -Original Message- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: Tuesday, April 21, 2009 8:38 AM To: Vemuri, Aparna Cc: r-help@r-project.org Subject: Re: [R] Fitting linear models On Apr 21, 2009, at 11:12 AM, Vemuri, Aparna wrote: David, Thanks for the suggestions. No, I did not label my dependent variable function. That was from my error in reading your call to lm. In my defense I am reasonably sure the proper assignment to arguments is lm(formula= ...) rather than lm(function= ...). My dependent variable PBW and all the independent variables are continuous variables. It is especially troubling since the order in which I input independent variables determines whether or not it gets a coefficient. Like I already mentioned, I checked the correlation matrix and picked the variables with moderate to high correlation with the independent variable. . So I guess it is not so naïve to expect a regression coefficient on all of them. Dimitri model1-lm(PBW~SO4+NO3+NH4), gives me the same result as before. Did you get the expected results with; model1-lm(formula=PBW~SO4+NO3+NH4+0) You could, of course, provide either the data or the results of str() applied to each of the variables and then we could all stop guessing. Aparna I am using the lm() function in R to fit a dependent variable to a set of 3 to 5 independent variables. For this, I used the following commands: model1-lm(function=PBW~SO4+NO3+NH4) Coefficients: (Intercept) SO4 NO3 NH4 0.01323 0.01968 0.01856 NA and model2-lm(function=PBW~SO4+NO3+NH4+Na+Cl) Coefficients: (Intercept) SO4 NO3 NH4 Na Cl -0.0006987 -0.0119750 -0.0295042 0.0842989 0.1344751 NA In both cases, the last independent variable has a coefficient of NA in the result. I say last variable because, when I change the order of the variables, the coefficient changes (see below). Can anyone point me to the reason R behaves this way? Is there anyway for me to force R to use all the variables? I checked the correlation matrices to makes sure there is no orthogonality between the variables. You really did not name your dependent variable function did you? Please stop that. Just a guess, ... since you have not provided enough information to do otherwise, ... Are all of those variables 1/0 dummy variables? If so and if you want to have an output that satisfies your need for labeling the coefficients as you naively anticipate, then put 0+ at the beginning of the formula or -1 at the end, so that the intercept will disappear and then all variables will get labeled as you expect. -- David Winsemius, MD Heritage Laboratories West Hartford, CT -- Dimitri Liakhovitski MarketTools, Inc. dimitri.liakhovit...@markettools.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.