Re: [R] ggplot2:

Hello, I'd like to resurrect this issue: is the varwidth equivalent (boxplot box-width scaling according to number of data points) emulatable in the 0.9.* versions of ggplot2? Width still doesn't seem capable of accepting a vector with length 1 ... Thank you for your input. Sincerely, Joh

[R] Examining how cases are similar by cluster, in cluster analysis

Hello, I used the following code to perform a cluster analysis on a dataframe consisting of 12 variables (coded as 1,0) and 63 cases. FS1 - read.csv(D://Arsontest2.csv,header=T,row.names=1) str(FS1) dmat - dist(FS1, method=binary) cl.test - hclust (dist(FS1, method =binary), ave)

Re: [R] help interpreting dudi.pco

I am new to R as well, it sounds like you would want to look at clustering, perhaps k-means clustering. Brian On Nov 18, 2012, at 12:19 AM, avadhoot velankar avadhoot.velan...@gmail.com wrote: I am working on morphometry of hairs and want to see if selected variables are giving

Re: [R] ggplot2: varwidth-equivalent for geom_boxplot?

Hello, I'd like to resurrect this issue: is the varwidth equivalent (boxplot box-width scaling according to number of data points) emulatable in the 0.9.* versions of ggplot2? Width still doesn't seem capable of accepting a vector with length 1 ... Thank you for your input. Sincerely, Joh

[R] centroid of hclust

Dear UseRs,i want to find centroid of clusters, which i generated by hclust. Is there a way doing that? i took mean to elements in each cluster to get centroid but i am not sure if i am right? thanks in advanceeliza [[alternative HTML version

[R] [lattice] how to label panels with variable value (not name)?

As described @ https://stat.ethz.ch/pipermail/r-help/2012-November/329442.html I now know how to prepare my netCDF data specifying gas concentrations over a 3D space (dimensions longitude, latitude, and (vertical) level), so as to call lattice::levelplot() like levelplot(conc ~ lon * lat |

Re: [R] polycor package

Dear Yves Many thanks for your quick reply, I really appreciate it! The reason I was working with a correlation matrix is because I wanted to calculate polychoric correlations first before submitting it to the actual sem command. I was unsure whether R would use polychoric correlations when

Re: [R] [lattice] how to label panels with variable value (not name)?

AFAICS, you will have to pre-compute the values and create a factor with levels these values as character strings. Then use this factor as the conditioning variable ?sprintf ?formatC ?round and the like may be useful to format the numerical values suitably for your labels. Cheers, Bert On Sun,

Re: [R] manipulating longitudinal data in r

At 08:56 17/11/2012, Kemi Racheal wrote: Dear list member, I have the following data example ke - data.frame(patid=c(1,1,1,2,3,3),a=c(1,2,2,1,1,2)) I want to add another variable b, such that the max of 'a' by id is returned i.e data ke becomes ke -

Re: [R] [lattice] how to label panels with variable value (not name)?

Tom Roche Tom_Roche at pobox.com writes: As described @ clipped However I will need to before-and-after compare this to the results of a reboxing, or 3D regridding, of this data, so I would prefer instead to label each panel in the lattice with the _value_ of the level (an

Re: [R] polycor package

The reason I was working with a correlation matrix is because I wanted to calculate polychoric correlations first before submitting it to the actual sem command. I was unsure whether R would use polychoric correlations when indicating which of the variables are ordered. A final question on the

Re: [R] manipulating longitudinal data in r

Michael, this comment doesn't seem appropriate to the question, since the sample data is a ragged array that requires the addition of NAs to fit into a wide format. --- Jeff NewmillerThe .

[R] list.files, recursively

Dear R developers, as far as I understand the manual of list.files(), there is only a way to exclude directories from the returned vector if you use list.files recursively. In non recursive mode, there seems to be no way of excluding directories (the include dirs argument does not seem to

[R] area under the curve

I have function y=x^5+x^8+ (cos(x))^3; I need to get random x; random y; and after that determine whether y below or above the given function. I tried a lot of version. but my code even doesnt workplz help with this expressions ;( I am a begiiner in R -- View this message in context:

[R] Fill in a pdf with colour

Good morning, I have a sampled probability density function and I have to fill in with colour a fraction of the total area beneath the curve (e.g. the 50% of the total area beneath the curve). I would like to know which is the simplest way to do this. Thank you for answering Best regards Nany23

Re: [R] [lattice] how to label panels with variable value (not name)?

Ken: I would have thought so, too. However: x - 1:10; y - runif(1:10); f - factor(rep(1:2,5)) xyplot(y~x|f, strip= function(which.panel,factor.levels,...){ lab - c(a,b)[which.panel] strip.default(factor.levels=lab,which.panel=which.panel,style=1,...)} ) produces label a for the

Re: [R] centroid of hclust

On 18.11.2012 14:25, eliza botto wrote: Dear UseRs,i want to find centroid of clusters, which i generated by hclust. Is there a way doing that? i took mean to elements in each cluster to get centroid but i am not sure if i am right? Yes, that is a common definition for centroid. Uwe

Re: [R] area under the curve

I mean i need to put random x in equation in order to determine value y. i used runifbut it doesn work... y= (runif(0,1))^5+(runif(0,1))^8+((cos runif(0,1)))^3; (=it doesnt work properly) now i have value y need to determine whether y is above or below my function. so after that

Re: [R] list.files, recursively

Hello, I believe that's a question for r-devel, but good point. It's docummented that in non-recursive calls to list.files subdirectory names are always included. (With a typo, There always are instead of They always are.) Rui Barradas Em 18-11-2012 17:20, Jannis escreveu: Dear R

Re: [R] Import excel file

On 16.11.2012 16:59, Rui Barradas wrote: Hello, I believe it is, but see package XLConnect. The vignette is very helpfull, with lots of examples. ... and many other ways. Please see the manual R Data Import/Excel that ships with R and has a whole section devoted to Excel. Best, Uwe

Re: [R] Fill in a pdf with colour

On 18.11.2012 13:02, nany23 wrote: Good morning, I have a sampled probability density function and I have to fill in with colour a fraction of the total area beneath the curve (e.g. the 50% of the total area beneath the curve). I would like to know which is the simplest way to do this. See

Re: [R] area under the curve

Which message are you referring to? Please quote. And please read the posting guide to the mailing list that asks you to do that! On 18.11.2012 18:55, Rlotus wrote: I mean i need to put random x in equation in order to determine value y. i used runifbut it doesn work... y=

Re: [R] area under the curve

On 18.11.2012 18:37, Rlotus wrote: I have function y=x^5+x^8+ (cos(x))^3; I need to get random x; random y; and after that determine whether y below or above the given function. I tried a lot of version. but my code even doesnt workplz help with this expressions ;( I am a begiiner in R

[R] i want to put the results of the list in a for loop

Dear R users, i want to put the results of a list to a for loop. i will give an example m1-matrix(rep(1,15),ncol=5)   ind.sgn - lapply(1:ncol(m1), combn, x = ncol(m1))  ind.sgn [[1]] [,1] [,2] [,3] [,4] [,5] [1,]    1    2    3    4    5 [[2]] [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]

Re: [R] tol in prcomp

On 17.11.2012 00:41, nmend wrote: Hi there, I was wondering if anyone could explain how you should set tol in the prcomp function. Using help(prcomp) explains it as a value indicating the magnitude below which components should be omitted. (Components are omitted if their standard deviations

Re: [R] [lattice] how to label panels with variable value (not name)?

Bert, A little bit of experimentation and I got this: library(lattice) x - 1:10 y - runif(1:10) f - factor(rep(1:2,5)) xyplot(y ~ x | f,  strip = function(which.panel, factor.levels,...){       lab - c(a,b) #[which.panel]       strip.default(factor.levels = lab,

Re: [R] vectorized plotmath expressions via substitute()

On 15.11.2012 10:52, Gabor Grothendieck wrote: On Wed, Nov 14, 2012 at 12:05 PM, Murat Tasan mmu...@gmail.com wrote: hi all - i've seen versions of this question before, but none seem to get directly at my solving my (probably very simple) issue: i simply want to annotate the tick marks on

Re: [R] discrete discriminant analysis

On 16.11.2012 16:05, Catarina Maia wrote: Hello, I am using the mda package and in particular the fda routine to classify in term of gear a set of 20 trips. I preformed a flexible discriminant analysis (FDA) using a set of 151 trips. FDAT1 - fda(as.factor(gear) ~ . , data =matrizR) A

Re: [R] additive interaction for a dichotomous dependent variable (i.e. risk difference)

On 16.11.2012 02:39, sarahw wrote: Did you ever get a response to this or resolve this yourself? You sent a message to the mailing list, but apparently not to the original poster. Uwe Ligges Many thanks! -- View this message in context:

Re: [R] [lattice] how to label panels with variable value (not name)?

Ken: Aha! Thank you. I was thrown by the phrase in the Help file, currently being drawn, which I parsed to the levels rather the conditioning variable. As I suspected, user error! Still not sure why strip.custom fails. -- Bert On Sun, Nov 18, 2012 at 10:25 AM, Ken Knoblauch

Re: [R] [lattice] how to label panels with variable value (not name)?

I looked at strip.custom()'s code and now get it, too. Here's the version for that: xyplot(y~x|f, strip=strip.custom(factor.levels=c(a,b),style=1)) Both of these require precomputing the factor.levels vector of character (or expression) labels. I see no obvious way to calculate them on the fly

Re: [R] [lattice] how to label panels with variable value (not name)?

OK -- to answer my own question below, here's how to compute them on the fly, which may be a better answer than what I gave in my earlier reply to Tom Roche: x - 1:10; y - runif(10); f - factor(rep(1:2,5)) xyplot(y~x|f, strip=function(which.panel,factor.levels,...){ lab -

Re: [R] i want to put the results of the list in a for loop

HI, Not sure if I understand it clearly. Are you looking for something like this: res1-list()  fun1-function(x,res){  for(i in 1:ncol(x)){  res[[i]]-list()  res[[i]]-combn(1:ncol(x),i)  }  res} fun1(m1,res1) [[1]] [,1] [,2] [,3] [,4] [,5] [1,]    1    2    3    4    5 [[2]] [,1]

Re: [R] R-Square in WLS

On Fri, Nov 16, 2012 at 4:48 PM, frespider frespi...@hotmail.com wrote: Hi, I am fitting a weighted least square regression and trying to compute SSE,SST and SSReg but I am not getting SST = SSReg + SSE and I dont know what I am coding wrong. Can you help please? For a start, you need to

[R] [lattice] format and rotation of strip text

Thanks to the lattice gurus on this list, and having reference to the excellent open-access Sarkar 2008 ISBN 978-0-387-75968-5 e-ISBN 978-0-387-75969-2 http://dx.doi.org/10.1007/978-0-387-75969-2 I now know how to label lattice panels by variable value: see thread starting @

Re: [R] Examining how cases are similar by cluster, in cluster analysis

If you just want a summary of the mean for each variable in each cluster, this will get you there: set.seed=42 FS1 - data.frame(matrix(sample(c(0, 1), 12*63, replace=TRUE), nrow=63, + ncol=12)) dmat - dist(FS1, method=binary) cl.test - hclust(dmat, method=average) plot(cl.test, hang=-1)

[R] identical matrices

Dear R users, I want to check matrices when i change the order of the rows or/and the order of the columns or/and the combination of them i will give an example what i want  1  -1  1  1  1  1   1  1 -1  -1 -1 -1    -1 -1  -1  -1   1  1    1  1 1 -1    1   1 these 2 matrices are

Re: [R] [lattice] how to label panels with variable value (not name)?

https://stat.ethz.ch/pipermail/r-help/2012-November/329450.html I [can now] prepare my netCDF data specifying gas concentrations over a 3D space (dimensions longitude, latitude, and (vertical) level), so as to call lattice::levelplot() like levelplot(conc ~ lon * lat | lev, data=data.frame)

Re: [R] Examining how cases are similar by cluster, in cluster analysis

David, Many thanks, I'm sure this will be helpful. What would also be helpful is if I can extract each cluster and examine id by variable, within the respective cluster. I could index the variables for each cluster and run such an analysis but thre must be a more efficient way of doing this

[R] Best prediction to use to use for basic problem?

I have a rather basic set of data. It is simply a variable that can be 0, 1 or 2 and its value over a series of time t0 - t9 like so: y: 1 1 2 0 1 2 2 1 2 1 x: t0 t1 t2 t3 t4 t5 t6 t7 t8

[R] How to fix it?

Gives a mistake that probability is not found x- runif(1,0,1) y-x^5+x^8+(cos(x))^3 yrandom- runif(10,0,3) if (yyrandom) {probabibility=y/yrandom} area=probability*3 -- View this message in context: http://r.789695.n4.nabble.com/How-to-fix-it-tp4649955.html Sent from the R help mailing list

[R] subtract multiple columns from single column for Nash Sutcliffe efficiency

Hi everyone, I am having trouble using my own data in the Nash-Sutcliffe efficiency (NSE) function. In R, this is what I have done: Vobsr - read.csv(Observed_Flow.csv, header = TRUE, sep =,) # see data below Vsimr - read.csv(1000Samples_Vsim.csv, header = TRUE, sep =,) # see data below Vobsr -

[R] Replace NA with something else

I am reading some data into R from an Excel spreadsheet using read.csv. Some of the original data that comes into column 1 from the spreadsheet is text that says NA. The NA stands for north america. As it comes in, R converts the NA over to NA. What is the cleanest way to change the NA values

[R] Decimal places

I am a beginner in programmin in general and R specifically. I would like to generate a set of random numbers in a normal distribution but to limit the decimal places in these numbers to only 2. I have been using x1 - runif(1,0,1) to generate my numbers. Can I add something to it to enable me

Re: [R] How to fix it?

On Sunday, November 18, 2012, Rlotus wrote: Gives a mistake that probability is not found x- runif(1,0,1) y-x^5+x^8+(cos(x))^3 yrandom- runif(10,0,3) if (yyrandom) {probabibility=y/yrandom} area=probability*3 You're ignoring the warning that's also given. Read ?ifelse and be enlightened

Re: [R] identical matrices

Please stop posting and re-posting the same questions. Michael On Sunday, November 18, 2012, Haris Rhrlp wrote: Dear R users, I want to check matrices when i change the order of the rows or/and the order of the columns or/and the combination of them i will give an example what i want

Re: [R] Decimal places

On Sunday, November 18, 2012, Rlotus wrote: I am a beginner in programmin in general and R specifically. I would like to generate a set of random numbers in a normal distribution but to limit the decimal places in these numbers to only 2. I have been using x1 - runif(1,0,1) to generate my

Re: [R] Replace NA with something else

use na.strings = '' x - read.table(text = 635 LA 201207 557329 + 636 LA 201208 683771 + 637 LA 201209 613851 + 638 LA 201210 764217 + 639 LA 201211 212897 + 782 NA 200701 875634 + 783 NA 200702 614856 + 784 NA 200703 521520 + 785 NA 200704 1406400, as.is =

Re: [R] Replace NA with something else

Hi, Try this: If you wanted to replace and from NA: dat1-read.table(text= 635    LA 201207  557329 636    LA 201208  683771 637    LA 201209  613851 638    LA 201210  764217 639    LA 201211  212897 782  NA 200701  875634 783  NA 200702  614856 784  NA 200703  521520 785  NA 200704 1406400

[R] Question about making histogram and x must be numeric

Hello all, I hope someone of you can help me out, I have searched other posts as well but I can't find any solution to the problem I'm dealing with. I want to make a histogram from the data Telephone Lines MDGdataset -read.csv(MDG_dataset_2010.csv, header=T) MDGdatasetAdapted -

Re: [R] Examining how cases are similar by cluster, in cluster analysis

Something like this? split(FS1, hcli8) $`1` X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 1 1 1 0 1 0 0 1 1 0 1 1 1 3 1 0 1 0 0 1 1 0 0 1 0 1 4 1 1 0 0 0 0 1 1 1 1 1 1 7 0 1 0 1 0 0 1 1 0 1 0 1 9 1 1 1 1 0 1 1 0 1 1 1 0

Re: [R] list.files, recursively

You can do the following to remove the directory names from the list: setdiff(list.files(full.names = TRUE), list.dirs(recursive = FALSE)) On Sun, Nov 18, 2012 at 12:20 PM, Jannis bt_jan...@yahoo.de wrote: Dear R developers, as far as I understand the manual of list.files(), there is only

Re: [R] Question about making histogram and x must be numeric

Most likely the value you are using in the 'hist' function is a factor. It would help if you included an 'str' of the object you are using. On Sun, Nov 18, 2012 at 4:45 PM, 9man lucas_9...@hotmail.com wrote: Hello all, I hope someone of you can help me out, I have searched other posts as well

Re: [R] Question about making histogram and x must be numeric

You tried setting shorter variable names, but the two lines of your data set that you gave us still have the original names, eg. Telephone.lines..per.100.people. -- David L Carlson Associate Professor of Anthropology Texas AM University College Station,

Re: [R] R-Square in WLS

Hi Thomas, Can you please edit my code in way it works. I appericated your help. I need to calculate the R2 in the Z scale and it is not make sense to me. or Can I have some documentation to read? Thanks Date: Sun, 18 Nov 2012 12:35:03 -0800 From: ml-node+s789695n4649969...@n4.nabble.com

Re: [R] Replace NA with something else

I inserted na.strings='' and that seemed to work except for a problem with a plot statement plot(x1$NA,type='l',ylab='M kg/ y ',xlab='') Error: unexpected numeric constant in plot(x1$NA tail(x1) AP EU LANA total Jun 2012 2.32 2.26 5.38 13.74 23.70 Jul 2012 2.46 2.21 5.33

Re: [R] additive interaction for a dichotomous dependent variable (i.e. risk difference)

In many situations the interactions indicated by additive risk models are false in the sense that they merely reflect restrictions so that risk are in [0,1]. I tend to prefer models that have no restrictions on the parameters. In fact, I think that a test of goodness of fit of a model can be

Re: [R] Replace NA with something else

Try x1$`NA` instead of x1$NA. HTH Quoting eric ericst...@aol.com: I inserted na.strings='' and that seemed to work except for a problem with a plot statement plot(x1$NA,type='l',ylab='M kg/ y ',xlab='') Error: unexpected numeric constant in plot(x1$NA tail(x1) AP EU LA

Re: [R] Examining how cases are similar by cluster, in cluster analysis

Hello David, Many thanks - this does exactly what I want and it lets me see whether the clusters make sense in terms of the patetrn of values where they join a cluster. Regards Bob Something like this? split(FS1, hcli8) $`1` X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 1 1 1 0 1 0

Re: [R] Best prediction to use to use for basic problem?

1. This is not really an R question. I suggest you post to a statistical list like stats.stackexchange.com instead. 2. I also suggest you consult a local statistician. You have a time series with a categorical response. That's not so simple as you think (and folks on stackexchange may provide you

Re: [R] [lattice] format and rotation of strip text

On Nov 18, 2012, at 12:42 PM, Tom Roche wrote: Thanks to the lattice gurus on this list, and having reference to the excellent open-access Sarkar 2008 ISBN 978-0-387-75968-5 e-ISBN 978-0-387-75969-2 http://dx.doi.org/10.1007/978-0-387-75969-2 I now know how to label lattice panels by

Re: [R] (no subject)

Sorry, got the answer (ticktype): persp(x, y, z1,box = TRUE, axes = TRUE,ticktype=detailed) Thanks! D. On Sun, Nov 18, 2012 at 8:20 PM, Dimitri Liakhovitski dimitri.liakhovit...@gmail.com wrote: Dear R-ers, I have 3 vectors - x, y, and z and want to plot a surface (z as the 3rd

[R] Error in `[.data.frame`... undefined columns selected

When I run this script on 9 variables, it works without problems. Z - data[,c(s1_1234_m,s2_1234_m,s3_1234_m,s4_1234_m,s5_1234_m,s6_1234_m,s7_1234_m,s8_1234_m,s9_1234_m )] However, when I run the script on 9 different variables, it does not work: Z -

Re: [R] [lattice] format and rotation of strip text

https://stat.ethz.ch/pipermail/r-help/2012-November/329479.html Hopefully [sufficiently] small, self-contained example: mailquotes omitted from start example to end example to ease rerunning the following code: # start example library(reshape2) library(lattice) lon=11 lat=7 lev=5

Re: [R] How to fix it?

On Mon, Nov 19, 2012 at 10:31 AM, R. Michael Weylandt michael.weyla...@gmail.com wrote: On Sunday, November 18, 2012, Rlotus wrote: Gives a mistake that probability is not found x- runif(1,0,1) y-x^5+x^8+(cos(x))^3 yrandom- runif(10,0,3) if (yyrandom) {probabibility=y/yrandom}

Re: [R] Error in `[.data.frame`... undefined columns selected

On Nov 18, 2012, at 6:34 PM, Eiko Fried wrote: When I run this script on 9 variables, it works without problems. Z - data [,c (s1_1234_m ,s2_1234_m ,s3_1234_m ,s4_1234_m ,s5_1234_m,s6_1234_m,s7_1234_m,s8_1234_m,s9_1234_m )] However, when I run the script on 9 different variables, it does not

Re: [R] [lattice] format and rotation of strip text

On Nov 18, 2012, at 7:05 PM, Tom Roche wrote: https://stat.ethz.ch/pipermail/r-help/2012-November/329479.html Hopefully [sufficiently] small, self-contained example: mailquotes omitted from start example to end example to ease rerunning the following code: # start example

Re: [R] ggplot2: varwidth-equivalent for geom_boxplot?

I'd like to resurrect this issue: is the varwidth equivalent (boxplot box-width scaling according to number of data points) emulatable in the 0.9.* versions of ggplot2? Width still doesn't seem capable of accepting a vector with length 1 ... No, and it's not currently on the to do list.

Re: [R] Basic Stats question

Hi -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-bounces@r- project.org] On Behalf Of Jeff Newmiller Sent: Saturday, November 17, 2012 3:21 PM To: Partha Sinha; r-help@r-project.org Subject: Re: [R] Basic Stats question If you read the posting guide

Re: [R] about lm

Hi Values ve or su came as factors during the input. Probably some mismatch in decimal point, nonumeric value somewhere, thousands separator or something else. At least str(mat) can help you (and us) what is the problem. Regards Petr -Original Message- From:

Re: [R] Sum Column in data.frame for Excel-Export

Hi maybe ?rbind Petr -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-bounces@r- project.org] On Behalf Of Mat Sent: Friday, November 16, 2012 3:42 PM To: r-help@r-project.org Subject: [R] Sum Column in data.frame for Excel-Export Hello together, I have

Re: [R] R-Square in WLS

On Nov 18, 2012, at 21:32 , Thomas Lumley wrote: On Fri, Nov 16, 2012 at 4:48 PM, frespider frespi...@hotmail.com wrote: Hi, I am fitting a weighted least square regression and trying to compute SSE,SST and SSReg but I am not getting SST = SSReg + SSE and I dont know what I am coding

[R] glht function in multcomp gives unexpected p=1 for all comparisons

Hi, I have data with binomial response variable (survival) and 2 categorical independent variables (site and treatment) (see below).  I have run a binomial GLM and found that both IVs and the interaction are significant.  Now I want to do a post-hoc test for all pairwise comparisons to see

[R] loop to subtract arrays / error

Hi everyone, I am having trouble with creating a loop to subtract arrays. In R, this is what I have done: Vobsr - read.csv(Observed_Flow.csv, header = TRUE, sep =,) # see data below Vsimr - read.csv(1000Samples_Vsim.csv, header = TRUE, sep =,) # see data below Vobsr - as.matrix(Vobsr[,-1])

Re: [R] Replace NA with something else

Hi, x1-read.table(text=  Year   AP  EU  LA  NA total Jun2012 2.32 2.26 5.38 13.74 23.70 Jul2012 2.46 2.21 5.33 12.94 22.94 Aug2012 2.69 2.24 5.28 13.32 23.54 Sep2012 2.62 2.28 5.14 12.99 23.06 Oct2012 2.61 2.27 5.31 12.59 22.80 Nov2012 2.55 2.18 5.08 12.56 22.39