[R] convering upper triangular matrix into vector
Hi I have two symmetrical distance matrices and want to compute the correlation coefficient between them (after turning them into vectors). Is there a way of selecting only the upper triangular part of each matrix, then convert this into a vector so I can compute the correlation? Many Thanks Eleni Rapsomaniki __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] convering upper triangular matrix into vector
[EMAIL PROTECTED] writes: Hi I have two symmetrical distance matrices and want to compute the correlation coefficient between them (after turning them into vectors). Is there a way of selecting only the upper triangular part of each matrix, then convert this into a vector so I can compute the correlation? X[upper.tri(X)] (as help.search(triangular) would have told you soon enough) -- 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 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] convering upper triangular matrix into vector
look at: '?upper.tri()' Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://www.med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: [EMAIL PROTECTED] To: R-help@stat.math.ethz.ch Sent: Sunday, October 02, 2005 11:04 AM Subject: [R] convering upper triangular matrix into vector Hi I have two symmetrical distance matrices and want to compute the correlation coefficient between them (after turning them into vectors). Is there a way of selecting only the upper triangular part of each matrix, then convert this into a vector so I can compute the correlation? Many Thanks Eleni Rapsomaniki __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] arima.sim bug?
Hi, I am using the arima.sim function to generate some AR time series. However, the function does not seem to produce exactly the same time series when I specify the innov parameter. For example r - rnorm(300) x - arima.sim(300, model=list(order=c(1,0,0),ar=c(.96)), innov=r, n.start=10) y - arima.sim(300, model=list(order=c(1,0,0),ar=c(.96)), innov=r, n.start=10) x[1:10] [1] 3.194806 4.214894 5.168017 7.925152 8.810817 9.131695 [7] 7.521283 8.266911 8.923429 9.651293 y[1:10] [1] -0.7202632 0.4564274 1.5598893 4.4613486 [5] 5.4855660 5.9394547 4.4567320 5.3249417 [9] 6.0991390 6.9399748 Given the fact that I have provided the innovations shouldn't the time series be exactly the same? Any help would be greatly appreciated. All the best, Sam. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] modeling language for optimization problems
Does anyone know whether R has its own modeling language for optimization problems (like SIMPLE in NuOPT for S-plus)? Paolo __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] arima.sim bug?
Kemp S E (Comp) wrote: Hi, I am using the arima.sim function to generate some AR time series. However, the function does not seem to produce exactly the same time series when I specify the innov parameter. For example r - rnorm(300) x - arima.sim(300, model=list(order=c(1,0,0),ar=c(.96)), innov=r, n.start=10) y - arima.sim(300, model=list(order=c(1,0,0),ar=c(.96)), innov=r, n.start=10) x[1:10] [1] 3.194806 4.214894 5.168017 7.925152 8.810817 9.131695 [7] 7.521283 8.266911 8.923429 9.651293 y[1:10] [1] -0.7202632 0.4564274 1.5598893 4.4613486 [5] 5.4855660 5.9394547 4.4567320 5.3249417 [9] 6.0991390 6.9399748 Given the fact that I have provided the innovations shouldn't the time series be exactly the same? Any help would be greatly appreciated. All the best, Sam. Not a bug, but a difference in seeds. Try: set.seed(1) r - rnorm(300) m - list(order = c(1,0,0), ar = c(.96)) set.seed(1) x - arima.sim(300, model = m, innov = r, n.start = 10) set.seed(1) y - arima.sim(300, model = m, innov = r, n.start = 10) all.equal(x, y) # [1] TRUE HTH, --sundar __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] plot question when type = b and pch is a vector
Dear tokas, How about: x - seq(0.01, 10, length = 20) xx - x[7] x[7] - NA plot(c(0, 10), c(-20, 20), type = n, xlab = x, ylab = expression(2 * alpha * log(x))) for(i in 1:4){ lines(x, 2 * i * log(x), lty = 1) text(xx, 2 * i * log(xx), i) } I hope this helps, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of toka tokas Sent: Sunday, October 02, 2005 5:37 AM To: r-help@stat.math.ethz.ch Subject: [R] plot question when type = b and pch is a vector Dear R users, I've been struggling some days with the following problem: I'm interesting in producing the following plot x - seq(0.01, 10, length = 20) plot(c(0, 10), c(-20, 20), type = n, xlab = x, ylab = expression(2 * alpha * log(x))) pch. - rep(NA, length(x)) for(i in 1:4){ pch.[7] - as.character(i) lines(x, 2 * i * log(x), type = b, pch = pch., lty = 1) } where all the line segments are connected, except from the 7th one where I've put the value of alpha -- in other words I'd like to produce a line plot where the label appears at each line with some white space around it. thanks in advance, tokas __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to get to the varable in a list
Lisa Wang wrote: Hello, I have a list lis as the following: $1 x1 x2 4 3 1 $2 x1 x2 3 3 2 5 3 2 $3 x1 x2 2 3 3 6 3 3 How do I get the x1 varible? for example ss1 I can only guess you mean something like lis[[1]][[x1]] Uwe Ligges This e-mail may contain confidential and/or privileged inf...{{dropped}} __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] User error (was arima.sim bug?)
On Sun, 2 Oct 2005, Kemp S E (Comp) wrote: Hi, I am using the arima.sim function to generate some AR time series. However, the function does not seem to produce exactly the same time series when I specify the innov parameter. For example r - rnorm(300) x - arima.sim(300, model=list(order=c(1,0,0),ar=c(.96)), innov=r, n.start=10) y - arima.sim(300, model=list(order=c(1,0,0),ar=c(.96)), innov=r, n.start=10) x[1:10] [1] 3.194806 4.214894 5.168017 7.925152 8.810817 9.131695 [7] 7.521283 8.266911 8.923429 9.651293 y[1:10] [1] -0.7202632 0.4564274 1.5598893 4.4613486 [5] 5.4855660 5.9394547 4.4567320 5.3249417 [9] 6.0991390 6.9399748 Given the fact that I have provided the innovations shouldn't the time series be exactly the same? No. Hint: where does the randomness for the burn-in come from? -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] modeling language for optimization problems
On Sun, 2 Oct 2005, Paolo Cavatore wrote: Does anyone know whether R has its own modeling language for optimization problems (like SIMPLE in NuOPT for S-plus)? No. Note that SIMPLE is the language of NUOPT, not of S-PLUS. There is an (extra-cost) interface module S+NUOPT, but it is an interface to NUOPT's engine. As far as I am aware R itself covers almost none of the ground of S+NUOPT, and available packages cover only a small part of it. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Size of jpegs/pngs
Dear all I have trouble with setting the size for jpegs and pngs. I need to save a dendrogram of 1000 words into a jpeg or png file. On one of my computers, the following works just fine: bb-agnes(aa, method=ward) jpeg(C:/Temp/test.txt, width=17000, height=2000) plot(bb) dev.off() On my main computer, however, this doesn't work: jpeg(C:/Temp/test.txt, width=17000, height=2000) Error in jpeg(C:/Temp/test.txt, width = 17000, height = 2000) : unable to start device devWindows In addition: Warning message: Unable to allocate bitmap This is a Windows XP Pro SP2 system, which is started with this chsort R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor1.1 year 2005 month06 day 20 language R which is started with a shortcut. C:\rw2011\bin\Rgui.exe --max-mem-size=1500M I checked the web and the R-help pages, tried out the ppsize option, and compared the options settings with those of the machine that works (which actually runs R 2.0.1 of 15 Nov 2004), but couldn't come up with an explanation. Any idea what I do wrong? STG -- Stefan Th. Gries Max Planck Inst. for Evol. Anthropology http://people.freenet.de/Stefan_Th_Gries Machen Sie aus 14 Cent spielend bis zu 100 Euro! Die neue Gaming-Area von Arcor - über 50 Onlinespiele im Angebot. http://www.arcor.de/rd/emf-gaming-1 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] p-level in packages mgcv and gam
Denis Chabot said the following on 2005-09-29 21:55: OK, I think I understand better but still have two points to clarify. The first one is about the number of df. I think those who replied on this objected to the way I chose df, not the fact that I would run a model with 7.4 df per se. If I read you correctly, I artificially reduce my p-value by using the estimated df found in a mgcv gam model into another model where I fix df. This is fine, I am quite willing not to run a second model with a fixed df and instead tell my readers that my model is marginally significant with a p-value of 0.03. This being said, do you know of guidelines for choosing df? A colleague told me he does not go above 10% of the number of points. Should I be concerned when mgcv estimates 7.4 df for 34 points? Note that for this particular model P 1e-16, and P is also very small if I fix df to either 4 or 7. My second point is the difference between models fitted by packages gam and mgcv. Sure, some of you have said different algorithms. And when Maybe that wasn't well put. Think of it as different implementations. The packages `mgcv' and `gam' are by no means the only implementations of GAM; see e.g. the `gss' package. I specify dfs, shouldn't P-values be very similar for the 2 packages? If not, what does it say of the confidence we can have in the models? Since there's no universally accepted definition of what a GAM is, you shouldn't expect the results to be the same. I draw your attention to this exampl: I obtained P-values of 0.50 and 0.03 with packages gam and mgcv respectively: In this case, however, you're trying to compare apples to oranges... library(gam) Loading required package: splines data(kyphosis) kyp1 - gam(Kyphosis ~ s(Number, 3), family=binomial, data=kyphosis) summary.gam(kyp1) Call: gam(formula = Kyphosis ~ s(Number, 3), family = binomial, data = kyphosis) Deviance Residuals: Min 1Q Median 3Q Max -1.3646 -0.6233 -0.4853 -0.3133 2.0965 (Dispersion Parameter for binomial family taken to be 1) Null Deviance: 83.2345 on 80 degrees of freedom Residual Deviance: 71.9973 on 76. degrees of freedom AIC: 79.9976 Number of Local Scoring Iterations: 7 DF for Terms and Chi-squares for Nonparametric Effects Df Npar Df Npar Chisq P(Chi) (Intercept) 1 s(Number, 3) 1 21.37149 0.50375 This test concerns only the non-linear part of the term s(Number, 3). In order to simultaneously test both the linear and non-linear part, as mgcv::summary.gam does, you'd kyp1.1 - gam(Kyphosis ~ 1, family=binomial, data=kyphosis) anova(kyp1.1, kyp1, test = Chisq) Analysis of Deviance Table Model 1: Kyphosis ~ 1 Model 2: Kyphosis ~ s(Number, 3) Resid. Df Resid. Dev Df Deviance P(|Chi|) 1 80. 83.234 2 76. 71.997 3.0001 11.237 0.011 HTH, Henric detach(package:gam) library(mgcv) This is mgcv 1.3-7 kyp2 - gam(Kyphosis ~ s(Number, k=4, fx=T), family=binomial, data=kyphosis) summary.gam(kyp2) Family: binomial Link function: logit Formula: Kyphosis ~ s(Number, k = 4, fx = T) Parametric coefficients: Estimate Std. Error z value Pr(|z|) (Intercept) -1.5504 0.3342 -4.64 3.49e-06 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Approximate significance of smooth terms: edf Est.rank Chi.sq p-value s(Number) 33 8.898 0.0307 * --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 R-sq.(adj) = 0.101 Deviance explained = 12.5% UBRE score = 0.075202 Scale est. = 1 n = 81 kyp2$deviance [1] 72.85893 kyp2$null.deviance [1] 83.23447 kyp2$df.null [1] 80 kyp2$df.residual [1] 77 How can we explain this huge difference? Denis Le 05-09-29 à 06:00, [EMAIL PROTECTED] a écrit : De : Henric Nilsson [EMAIL PROTECTED] Date : 29 septembre 2005 03:55:19 HAE À : [EMAIL PROTECTED] Cc : r-help@stat.math.ethz.ch Objet : Rép : [R] p-level in packages mgcv and gam Répondre à : Henric Nilsson [EMAIL PROTECTED] Yves Magliulo said the following on 2005-09-28 17:05: hi, i'll try to help you, i send a mail about this subject last week... and i did not have any response... I'm using gam from package mgcv. 1) How to interpret the significance of smooth terms is hard for me to understand perfectly : using UBRE, you fix df. p-value are estimated by chi-sq distribution using GCV, the best df are estimated by GAM. (that's what i want) and p-values This is not correct. The df are estimated in both cases (i.e. UBRE and GCV), but the scale parameter is fixed in the UBRE case. Hence, by default UBRE is used for family = binomial or poisson since the scale parameter is assumed to be 1. Similarly, GCV is the default for family = gaussian since we most often want the scale (usually denoted sigma^2) to be
Re: [R] Memory management on Windows (was Size of jpegs/pngs)
I think this an issue about the amount of graphics memory. You are asking for an image of about 17*2*3 = 102Mb, and you need more than that. From the help page: Windows imposes limits on the size of bitmaps: these are not documented in the SDK and may depend on the version of Windows. It seems that 'width' and 'height' are each limited to 2^15-1 and there is a 16Mb limit on the total amount of memory in Windows 95/98/ME. so I do wonder why you are surprised. My laptop appears to be limited to about half your example with a 128Mb graphics card (and lots of other things going on). On Sun, 2 Oct 2005 [EMAIL PROTECTED] wrote: Dear all I have trouble with setting the size for jpegs and pngs. I need to save a dendrogram of 1000 words into a jpeg or png file. On one of my computers, the following works just fine: bb-agnes(aa, method=ward) jpeg(C:/Temp/test.txt, width=17000, height=2000) plot(bb) dev.off() On my main computer, however, this doesn't work: jpeg(C:/Temp/test.txt, width=17000, height=2000) Error in jpeg(C:/Temp/test.txt, width = 17000, height = 2000) : unable to start device devWindows In addition: Warning message: Unable to allocate bitmap This is a Windows XP Pro SP2 system, which is started with this chsort R.version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major2 minor1.1 year 2005 month06 day 20 language R which is started with a shortcut. C:\rw2011\bin\Rgui.exe --max-mem-size=1500M I checked the web and the R-help pages, tried out the ppsize option, and compared the options settings with those of the machine that works (which actually runs R 2.0.1 of 15 Nov 2004), but couldn't come up with an explanation. Any idea what I do wrong? Did you read the help page? -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] rate instead of scale in ?ks.test
I am not sure whether I'm doing something wrong or there is a bug in the documentation of ks.test. Following the posting guide, as I'm not sure, I haven't found this in the bug tracker, and the R FAQ says that stats is an Add-on package of R, I think this is the place to send it. ?ks.test provides the example QUOTE # Does x come from a shifted gamma distribution with shape 3 and scale 2? ks.test(x+2, pgamma, 3, 2) # two-sided /QUOTE Maybe it should say ``with shape 3 and rate 2''? If I do x - rgamma(1e6, shape = 3, scale = 2) ks.test( x, 'pgamma', 3, 2 ) One-sample Kolmogorov-Smirnov test data: x D = 0.7513, p-value 2.2e-16 alternative hypothesis: two.sided whereas with y - rgamma(1e6, shape = 3, rate = 2) ks.test( y, 'pgamma', 3, 2 ) One-sample Kolmogorov-Smirnov test data: y D = 5e-04, p-value = 0.9469 alternative hypothesis: two.sided If forcing the parameter meaning: ks.test( x, 'pgamma', shape = 3, scale = 2 ) One-sample Kolmogorov-Smirnov test data: x D = 9e-04, p-value = 0.3645 alternative hypothesis: two.sided version platform i386-pc-linux-gnu arch i386 os linux-gnu system i386, linux-gnu status major2 minor1.1 year 2005 month06 day 20 language R packageDescription( stats ) Package: stats Version: 2.1.1 Priority: base Title: The R Stats Package Author: R Development Core Team and contributors worldwide Maintainer: R Core Team [EMAIL PROTECTED] Description: R statistical functions License: GPL Version 2 or later. Built: R 2.1.1; i386-pc-linux-gnu; 2005-06-29 23:30:55; unix -- File: /usr/lib/R/library/stats/DESCRIPTION Cheers, -- Ramón Casero Cañas web:http://www.robots.ox.ac.uk/~rcasero/ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] p-level in packages mgcv and gam
Thank you very much Henric, now I see! Denis Le 05-10-02 à 11:47, Henric Nilsson a écrit : This test concerns only the non-linear part of the term s(Number, 3). In order to simultaneously test both the linear and non-linear part, as mgcv::summary.gam does, you'd kyp1.1 - gam(Kyphosis ~ 1, family=binomial, data=kyphosis) anova(kyp1.1, kyp1, test = Chisq) Analysis of Deviance Table Model 1: Kyphosis ~ 1 Model 2: Kyphosis ~ s(Number, 3) Resid. Df Resid. Dev Df Deviance P(|Chi|) 1 80. 83.234 2 76. 71.997 3.0001 11.237 0.011 HTH, Henric __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] generalized linear model for multinomial data?
Dear All: Does R have the package as in SAS's generalized logits model for nominal response data? I have searched but cannot find the windows package. Many thanks, HS __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] question for some tests
Hi! I come from Poland, and my English is not as good as it should be, so I have to apologize for my mistakes. I would like to ask you for a package, where I can find a test, which is named The Test Of Series (in Poland). In econometric models we use this test to check if the residuals are completely random ( I mean with destiny). And one more thing - I would like to know where I can find a test for symetric of residuals (called The Test For The Indicator Of The Structure). And In general, where I can the list for the tests being used in ecometrics. With greetings, Krzysztof Sroka Please, answer on my e-mail address: [EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] question for some tests
On Sun, 2 Oct 2005, Krzysztof Sroka wrote: Hi! I come from Poland, and my English is not as good as it should be, so I have to apologize for my mistakes. I would like to ask you for a package, where I can find a test, which is named The Test Of Series (in Poland). In econometric models we use this test to check if the residuals are completely random ( I mean with destiny). And one more thing - I would like to know where I can find a test for symetric of residuals (called The Test For The Indicator Of The Structure). And In general, where I can the list for the tests being used in ecometrics. R has Task Views, posted on the Archive Network (CRAN), and one of these is for econometrics: http://cran.r-project.org/src/contrib/Views/Econometrics.html It is quite dense (many links), certainly has what you need, but you will have to look around to find what fits your needs more closely. I would look at the lmtest package to start with, the Task View includes much more. With greetings, Krzysztof Sroka Please, answer on my e-mail address: [EMAIL PROTECTED] [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Access to particular predict value
If you would still like help from this list, please provide a very simple reproducible example to help what you tried that didn't quite work. There are several functions that use inverse distance weighting (IDW) and kriging for geostatistical computations. spencer graves p.s. PLEASE do read the posting guide! www.R-project.org/posting-guide.html. I believe that people who use the posting guide generally get better answers quicker, because it is easier for others to understand what they want. Toni Viúdez wrote: Hi everybody: I just generate interpolation maps with differents methods, like IDW and kriging, and now i want to compare the predict values versus real, and i don't who is the commant to do it. I want for example, if I pass from console the coordinates of a place, return me the predict value. Could anybody tell me how should do?. Thanks in advance -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] generalized linear model for multinomial data?
Dear Hongyu, See multinom() in the nnet package, associated with Venables and Ripley's Modern Applied Statistics with S. John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Hongyu Sun Sent: Sunday, October 02, 2005 3:07 PM To: r-help@stat.math.ethz.ch Subject: [R] generalized linear model for multinomial data? Dear All: Does R have the package as in SAS's generalized logits model for nominal response data? I have searched but cannot find the windows package. Many thanks, HS __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R-code for binormla distribution
What are you trying to do that requires binormal probabilities other than pmvnorm? spencer graves Nabil Channouf wrote: Dear users, does any one have a code (S or R) to compute the binormal distribution (or the upper its quadrant area) other than the pmvnorm. Thanks -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] generalized linear model for multinomial data?
Dear Professor: Thanks! I just found it. It is bundled within the VR package. Hongyu - Original Message - From: John Fox [EMAIL PROTECTED] To: 'Hongyu Sun' [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Sent: Sunday, October 02, 2005 3:27 PM Subject: RE: [R] generalized linear model for multinomial data? Dear Hongyu, See multinom() in the nnet package, associated with Venables and Ripley's Modern Applied Statistics with S. John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Hongyu Sun Sent: Sunday, October 02, 2005 3:07 PM To: r-help@stat.math.ethz.ch Subject: [R] generalized linear model for multinomial data? Dear All: Does R have the package as in SAS's generalized logits model for nominal response data? I have searched but cannot find the windows package. Many thanks, HS __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Selecting columns of a table (not consecutive)
Hi everyone, I have a table with 10 columns and many rows. I want to select the 3rd,4th and 10th column. If I wanted to select the 3rd and 4th it would be no problem. q3[ ,(3:4)] but how do I select the 10th along with it? Thanks for your help, George p.s Please respond if you know, assignment is due tomorrow... Thanks again.. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Selecting columns of a table (not consecutive)
George Papayiannis wrote: Hi everyone, I have a table with 10 columns and many rows. I want to select the 3rd,4th and 10th column. If I wanted to select the 3rd and 4th it would be no problem. q3[ ,(3:4)] but how do I select the 10th along with it? Thanks for your help, George p.s Please respond if you know, assignment is due tomorrow... Thanks again.. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Hi George, Try subset(q3, select=c(3,4,10)) Nolwenn __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] What is Mandel's (Fitting) Test?
Hello everyone, A little background first: I have collected psychophysical data from 12 participants, and each participant's data is represented as a scatter plot (Percieved roughness versus Physical roughness). I would like to know whether, on average, this data is best fit by a linear function or by a quadratic function. (we have a priori reasons to expect a quadratic) Some of my colleagues have suggested the following methods of testing this: 1. For each participant, calculate the r-square values for linear and quadratic fits, z-transform the resulting values. Collect these z-transformed scores and then perform a dependent t-test across participants. If significant, then a quadratic fits better. 2. For each participant, calculate the amount of variance left over from the linear fit that is accounted for by the quadratic fit. Perform a one-sample t-test to see if this population of scores differs from zero 3. Same as #2, but z-transform before performing the t-test. However, I'm sure that these tests fail to take into account the fact that a quadratic function will generally have an advantage over a linear function simply by dint of having more terms to play with. So I've been looking for a test that takes this advantage into account and I came across something called the Mandel Test. It is available in the quantchem package, but the manual contains a very meagre description of it's details (assumptions, etc). Furthermore, besides biology/chemistry papers that reference it in passing, I've been able to find only one reference online that addresses it's use (http://www.econ.kuleuven.be/public/ndbae06/PDF-FILES/vanloco.doc), but even then it lacks specificity. So the question is, what is the Mandel Test? What are the assumptions and limitations of the test? Does it sound appropriate for my purposes (if not, how about the other tests suggested above)? How does it differ from the Lack-of-Fit test? Any help would be greatly appreciated. Cheers, Mike -- Mike Lawrence, BA(Hons) Research Assistant to Dr. Gail Eskes Dalhousie University QEII Health Sciences Centre (Psychiatry) [EMAIL PROTECTED] The road to Wisdom? Well, it's plain and simple to express: Err and err and err again, but less and less and less. - Piet Hein __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] R-code for binormla distribution
Why do you say we do not have the package that contains pmvnorm? It is in library(mvtnorm), which you can get via CRAN (the Comprehensive R Archive Network). If you are using Windows or a standard Linux of MacOS, install.packages('mvtnorm') should install it on your computer (unless you are using a computer with special protection that makes this difficult; if you are using XEmacs, you should exit R under XEmacs and run install.packages from Rgui). Then library('mvtnorm') should make it available to your R session. If you are using some other operating system, you may need to access the source, but that's available via CRAN alse. If you have not tried install.packages, please do so. If you have tried install.packages and can't make it work, please describe your operating system, etc., as explained in the posting guide! www.R-project.org/posting-guide.html. spencer graves Nabil Channouf wrote: Dear Mr. Graves, i'm trying to write my own function with R, because we do not have the package that contains pmvnorm and i need to compute the upper quadrant area of the bivariate standard normal, or at least to know the details of the function pmvnorm written in Splus or in R. Thanks Spencer Graves wrote: What are you trying to do that requires binormal probabilities other than pmvnorm? spencer graves Nabil Channouf wrote: Dear users, does any one have a code (S or R) to compute the binormal distribution (or the upper its quadrant area) other than the pmvnorm. Thanks -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Local install of a contributed package under Linux (was: R-code for binormla distribution)
There probably is a way under Linux to install a contributed package in your local directory. I don't know how, but if you read the documentation on download.packages and install.packages, you might be able to do it without bothering your Info Tech department. I know there is a way, and with luck someone who knows will respond to this email. If you try something with download.packages and install.packages, and you can't make it work, please tell us what you try and what doesn't work. Maybe someone who uses R on Linux will respond. If not, ask your Info Tech department. Spencer Graves Nabil Channouf wrote: Mr Graves, we are working with linux and it is so complicated to install any thing, it should be via the informaticians of the department and we have to make a request and wait. I will ask them to do it on monday. Thank you so much Spencer Graves wrote: Why do you say we do not have the package that contains pmvnorm? It is in library(mvtnorm), which you can get via CRAN (the Comprehensive R Archive Network). If you are using Windows or a standard Linux of MacOS, install.packages('mvtnorm') should install it on your computer (unless you are using a computer with special protection that makes this difficult; if you are using XEmacs, you should exit R under XEmacs and run install.packages from Rgui). Then library('mvtnorm') should make it available to your R session. If you are using some other operating system, you may need to access the source, but that's available via CRAN alse. If you have not tried install.packages, please do so. If you have tried install.packages and can't make it work, please describe your operating system, etc., as explained in the posting guide! www.R-project.org/posting-guide.html. spencer graves Nabil Channouf wrote: Dear Mr. Graves, i'm trying to write my own function with R, because we do not have the package that contains pmvnorm and i need to compute the upper quadrant area of the bivariate standard normal, or at least to know the details of the function pmvnorm written in Splus or in R. Thanks Spencer Graves wrote: What are you trying to do that requires binormal probabilities other than pmvnorm? spencer graves Nabil Channouf wrote: Dear users, does any one have a code (S or R) to compute the binormal distribution (or the upper its quadrant area) other than the pmvnorm. Thanks -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Local install of a contributed package under Linux (was: R-code for binormla distribution)
Try R CMD INSTALL -l lib pkgs The help file is in the utils package. I'm sure this is documented in the manual too. Jon -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page: http://www.sas.upenn.edu/~baron __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] grob questions
If I run the following example from: http://www.stat.auckland.ac.nz/~paul/grid/doc/grobs.pdf grid.newpage() pushViewport(viewport(w = 0.5, h = 0.5)) myplot - gTree(name = myplot, children = gList(rectGrob(name = box, + gp = gpar(col = grey)), xaxisGrob(name = xaxis))) grid.draw(myplot) grid.edit(myplot::xaxis, at = 1:10/11) grid.edit(myplot::xaxis::labels, label = round(1:10/11, 2)) grid.edit(myplot::xaxis::labels, y = unit(-1, lines)) then str(myplot$children$xaxis) lists 'at' but not the 'labels'. yet if I do this then the labels are listed: xx - xaxisGrob(name = myX, at = 1:10) childNames(xx) [1] major ticks labels 1. How do I get to labels in the first case? 2. Is there a better construct than myplot$children$xaxis? Thanks. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html