Re: [R] nonlinear modeling with rational functions?

On Wed, Jun 16, 2004 at 10:28:39AM -0500, Spencer Graves wrote: Rational functions (ratios of polynomials) often provide good approximations to many functions. Does anyone know of any literature on nonlinear modeling with rational functions, sequential estimation, diagnostics, etc.?

[R] PCA and AR on circular data

Could anyone advise the best method of performing PCA and/or AR modelling on circular data? Until now I have been working with vectors (speed*direction), but am now wanting to look purely at the directional aspect. Thanks in advance.. Laura __ [EMAIL

[R] Date Calculation

Hi, I've been playing with: joinDate - format(strptime(as.vector(forum[,2]), %d-%b-%y), +%d-%b-%Y) today - format(strptime(as.vector(14-Jun-04), %d-%b-%Y), + %d-%b-%Y) joinDate [1] 04-Feb-2004 13-Feb-2004 26-Feb-2004 27-Feb-2004 27-Feb-2004 [6] 27-Feb-2004

Re: [R] subset(..., drop=TRUE) doesn't seem to work.

Hoi Peter, --On woensdag 16 juni 2004 17:35 +0200 Peter Dalgaard [EMAIL PROTECTED] wrote: Anyways, the way out is d2 - subset(dd,c==1) ifac - sapply(dd,is.factor) d2[ifac] - lapply(d2[ifac],factor) or d2 - subset(dd,c==1) d2[] - lapply(d2, function(x) if (is.factor(x)) factor(x) else x) My

Re: [R] nonlinear modeling with rational functions?

Tamas == Tamas Papp [EMAIL PROTECTED] on Thu, 17 Jun 2004 08:40:13 +0200 writes: Tamas On Wed, Jun 16, 2004 at 10:28:39AM -0500, Spencer Graves wrote: Rational functions (ratios of polynomials) often provide good approximations to many functions. Does anyone know of any

Re: [R] Date Calculation

KKWa == Ko-Kang Kevin Wang [EMAIL PROTECTED] on Thu, 17 Jun 2004 19:57:39 +1200 writes: KKWa Hi, KKWa I've been playing with: joinDate - format(strptime(as.vector(forum[,2]), %d-%b-%y), KKWa +%d-%b-%Y) today - format(strptime(as.vector(14-Jun-04),

Re: [R] Date Calculation

You have formatted the dates, so you are trying to subtract character strings. Convert to Date instead (although you can subtract POSIXlt dates). On Thu, 17 Jun 2004, Ko-Kang Kevin Wang wrote: Hi, I've been playing with: joinDate - format(strptime(as.vector(forum[,2]), %d-%b-%y), +

Re: [R] Tutorial for graphics

On Wed, Jun 16, 2004 at 10:12:49PM -0400, Phil wrote: I'm coming to R from Matlab and I'm finding it difficult to find a good introduction to graphics in R. (The best I've found so far is Ch. 4 of Emmanuel Paradis R for Beginners. Still, I have been unable to discover simple things like how

RE: [R] Date Calculation

Hi, -Original Message- From: Martin Maechler [mailto:[EMAIL PROTECTED] Kevin, [[did you have tough day? usually your Q/A are much better ;-()]] Thanks to those who have replied, and yes shame on me.. [I also realised I can just use Sys.Date() to get today's date, instead

[R] packages data-sets

It's possible to create a package with functions and data, from which the use library(pkg-name) attaches not only the functions, but also the data? I want avoid to use data(dataset, package=name) because this makes a global copy of the data-set ... Anyone could help me? Meinhard

Re: [R] packages data-sets

On Thu, 17 Jun 2004, Meinhard Ploner wrote: It's possible to create a package with functions and data, from which the use library(pkg-name) attaches not only the functions, but also the data? I want avoid to use data(dataset, package=name) because this makes a global copy of the

Re: [R] Question : simtest result

Dear Mr.Torsten: At Wed, 16 Jun 2004 08:04:08 +0200 (CEST), Torsten Hothorn wrote: Call: simtest.formula(formula = y4 ~ f2, data = dat2, type = Dunnett) Dunnett contrasts for factor f2 Contrast matrix: f21 f22 f23 f24 f25 f22-f21 0 -1 1 0 0 0 f23-f21 0

[R] How to order a vector

Hi all I have a vector like this 2003 2002 2001 2000 1999 1998 1997 1996 106 105 106 106105 106 101 107 How can I get it sorted right(19962003)? Thank you Luis Ridao Cruz Fiskirannsóknarstovan Nóatún 1 P.O. Box 3051 FR-110 Tórshavn Faroe Islands Phone: +298

Re: [R] How to order a vector

[EMAIL PROTECTED] wrote on 17/06/2004 13:10:35: Hi all I have a vector like this 2003 2002 2001 2000 1999 1998 1997 1996 106 105 106 106105 106 101 107 How can I get it sorted right(19962003)? rev(1:4) [1] 4 3 2 1 HTH, Tobias

Re: [R] How to order a vector

Hi! I assume that 2003 2002 are the names of your vector myvector myvector[order(names(myvector))] Eryk *** REPLY SEPARATOR *** On 6/17/2004 at 12:10 PM Luis Rideau Cruz wrote: Hi all I have a vector like this 2003 2002 2001 2000 1999 1998 1997 1996 106 105 106 106

Re: [R] How to order a vector

See ?order. In particular, if you do: a - order(names(vec)), a will contain the indices of the vector in order, so vec[a] will be in the correct order. Sean On 6/17/04 7:10 AM, Luis Rideau Cruz [EMAIL PROTECTED] wrote: Hi all I have a vector like this 2003 2002 2001 2000 1999 1998 1997

[R] Using predict.lm()

Greetings, Following the example in help(predict.lm): x - rnorm(15) y - x + rnorm(15) new - data.frame(x = seq(-3, 3, 0.5)) predict(lm(y ~ x), new) predicts the response elements corresponding to new$x as can be viewed by: plot(x,y) lines(new$x,predict(lm(y ~ x),

Re: [R] Using predict.lm()

On Thu, 17 Jun 2004, Steven White wrote: Following the example in help(predict.lm): x - rnorm(15) y - x + rnorm(15) new - data.frame(x = seq(-3, 3, 0.5)) predict(lm(y ~ x), new) predicts the response elements corresponding to new$x as can be viewed by:

Re: [R] non-linear binning? power-law in R

Why not try to avoid binning (and density plot) at all? An alternative could be a qqplot (as a log-log-plot), e.g. plot(ppoints(length(x4)), x4[order(x4)], log=xy) abline(lm(log(x4[order(x4)])~log(ppoints(length(x4, col=red) If the assumptions of uniform distribution and power

Re: [R] Resolution of plots

On Wed, 16 Jun 2004, Prof Brian Ripley wrote: You will have to tell us more. Exporting how: to what format using what device and what exact command on what operating system? The only device I know of that even knows about dpi is bitmap() and that has no such limit unless imposed by your

[R] 2D Kolmogorov-Smirnov test: solution

Hi - A little while ago I posted a question about the implementation of a two-dimensional analog of the Kolmogorov-Smirnov test in R[1][2]. As there isn't one, as far as I know, people might be interested in a very fast C++ implementation called MUAC which is available as a function and as

[R] Re: Clustering in R

Thanks a lot, Michael! I cc to R-help, where this question really belongs {as the 'Subject' suggests itself...} -- please drop 'bioconductor' from CC'ing further replies. michael == michael watson (IAH-C) [EMAIL PROTECTED] on Thu, 17 Jun 2004 09:16:59 +0100 writes: michael OK,

[R] Error with arima()

Could someone please give a brief explanation, or pointer to an explanation, of the following error: arima(ts.growth, order = c(1,0,0),include.mean=T) Error in arima(ts.growth, order = c(1, 0, 0), include.mean = T) : non-stationary AR part from CSS and why it does not arise with

[R] disappointed with x-axes in hist and density plots

I've got a few issues with the x-axes in the histogram and density plots. First, often the default x-axis doesn't even extend to the length of my data. R often draws histogram bars (or density lines) farther than the drawn x-axis extends. For example, I might have a histogram bar at -15,000. But

[R] L é vy distribution?

Hi all, Does R support the Lévy distribution? I looked everywhere and couldn't find it. Thanks for any help! Gabriel __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide!

Re: [R] disappointed with x-axes in hist and density plots

On Thu, 17 Jun 2004, Rishi Ganti wrote: I've got a few issues with the x-axes in the histogram and density plots. First, often the default x-axis doesn't even extend to the length of my data. R often draws histogram bars (or density lines) farther than the drawn x-axis extends. For

Re: [R] L é vy distribution?

Gabriel Erbano wrote: Hi all, Does R support the Lévy distribution? I looked everywhere and couldn't find it. Thanks for any help! Gabriel I believe Jim Lindsey's rmutil package supports this though I've never used it: http://popgen0146uns50.unimaas.nl/~jlindsey/rcode.html BTW, I found this out

Re: [R] Error with arima()

It means that the CSS estimates from the model are invalid. That's possible as CSS does not enforce validity. It probably means the model is inappropriate. On Thu, 17 Jun 2004, Dan Bebber wrote: Could someone please give a brief explanation, or pointer to an explanation, of the following

[R] R help in Firefox on Windows XP

I had to reinstall my machine, so I installed Firefox 0.9 as browser I am using WinXP and R 1.9.1 beta. Now search in R html help does not work. I checked that the Java VM is working correctlt, Sun's test site says my installation is OK. Firefoxalso tells me that Applet Searchengine loaded Applet

[R] beta regression in R

Hello, I'm using optim to program a set of mle regression procedures for non-normal disturbances. This is for teaching and expository purposes only. I've successfully programmed the normal, generalized gamma, gamma, weibull, exponential, and lognormal regression functions. And optim returns

Re: [R] L é vy distribution?

Actually, I tried searching the R site, but it just returned 4 hits, and one of them was about Lindsey's package. But since it was not listed in the packages section, I thought it was broken... Well, my bad! Thanks! From: Sundar Dorai-Raj [EMAIL PROTECTED] Organization: PDF Solutions, Inc.

[R] Question on lists and vectors of lists

I have an elementary programming question. Could someone please point me in the right direction? I have a function which will run for thousands of companies. At each invocation, it returns 2 numbers. I plan to do something like: think_one_firm - function(filename) { # Do stuff

Re: [R] disappointed with x-axes in hist and density plots

TL == Thomas Lumley [EMAIL PROTECTED] on Thu, 17 Jun 2004 09:53:33 -0700 (PDT) writes: TL On Thu, 17 Jun 2004, Rishi Ganti wrote: I've got a few issues with the x-axes in the histogram and density plots. First, often the default x-axis doesn't even extend to the length

Re: Re: [R] disappointed with x-axes in hist and density plots

Thanks, but even with axis() I can't get the x-axis to extend to the sides. Try, e.g., x = rnorm(1000) you should have some values in excess of 3 (or below -3). I want to draw the x-axis from -4 to 4, thus encapsulating all points. axis(1,-4:4) but it won't draw. It TRIES to draw it, but I

Re: [R] R help in Firefox on Windows XP

On Thu, 2004-06-17 at 12:06, Erich Neuwirth wrote: I had to reinstall my machine, so I installed Firefox 0.9 as browser I am using WinXP and R 1.9.1 beta. Now search in R html help does not work. I checked that the Java VM is working correctlt, Sun's test site says my installation is OK.

Re: [R] Question on lists and vectors of lists

Try this : silly.fn - function(x){ return( c(x^2, x^3) ) } output - matrix( nr=100, nc=2 ) for(i in 1:100){ my.x - rnorm(1) output[i, ] - silly.fn( my.x ) rownames(output)[i] - paste(Dataset, i, sep=) } colnames(output) - c(squared, cubed) Notice that silly.fn returns a vector which is set

Re: [R] Question on lists and vectors of lists

On Thu, 17 Jun 2004 22:56:29 +0530 Ajay Shah wrote: I have an elementary programming question. Could someone please point me in the right direction? I have a function which will run for thousands of companies. At each invocation, it returns 2 numbers. I plan to do something like:

Re: [R] R help in Firefox on Windows XP

Erich Neuwirth wrote: I had to reinstall my machine, so I installed Firefox 0.9 as browser I am using WinXP and R 1.9.1 beta. Now search in R html help does not work. A workaround (which is slow, provisional, and as yet untested on your configuration) is to use a different help engine,

[R] problem with restore and some .RData

Hi, I have problem with the restore function in some .RData using R 1.9.0 Look the error: [EMAIL PROTECTED] RAnalise]$ R ... Error: object 'family' not found whilst loading namespace 'MASS' Fatal error: unable to restore saved data in .RData But if I load this .RData with the load() function

RE: Re: [R] disappointed with x-axes in hist and density plots

You haven't read ?axis, I guess. Try using the at= argument. Whatever you do not like about the default output, chances are you can customize it to your heart's content... Andy From: Rishi Ganti Thanks, but even with axis() I can't get the x-axis to extend to the sides. Try, e.g.,

Re: [R] disappointed with x-axes in hist and density plots

On Thu, 17 Jun 2004 10:28:57 -0700 Rishi Ganti wrote: Thanks, but even with axis() I can't get the x-axis to extend to the sides. Try, e.g., x = rnorm(1000) you should have some values in excess of 3 (or below -3). I want to draw the x-axis from -4 to 4, thus encapsulating all

Re: [R] disappointed with x-axes in hist and density plots

x - rnorm(1000) y - rnorm(1000) plot(x,y) axis(1,-4,4) (speculation that attempting above ... not what you want to do...) rather do plot(x,y,xlim = c(-4,4)) Rishi Ganti wrote: Thanks, but even with axis() I can't get the x-axis to extend to the sides. Try, e.g., x = rnorm(1000) you should

Re: [R] R help in Firefox on Windows XP

Works for me with Firefox 0.8. James W. MacDonald Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 Prof Brian Ripley [EMAIL PROTECTED] 06/17/04 02:10PM Works for me. Did it work with Firefox 0.8?

Re: [R] disappointed with x-axes in hist and density plots

Rishi Ganti [EMAIL PROTECTED] writes: Thanks, but even with axis() I can't get the x-axis to extend to the sides. Try, e.g., x = rnorm(1000) you should have some values in excess of 3 (or below -3). I want to draw the x-axis from -4 to 4, thus encapsulating all points.

RE: [R] nlme graphics in a loop problem

Could it be that you need print(plot(model)) inside the loop? I believe plot() methods in nlme are mostly lattice graphics, which needs to be explicitly print()ed inside functions and loops. Andy From: Chris Knight Hi, I'm fitting mixed effects models using the lme function of the nlme

Re: [R] nlme graphics in a loop problem

yes, use the print command, and if you can save each plot in the loop, that will work. I had to do this for a loop and tables I was making. I had to save the first table to a matrix, then bind each successive table to that matrix, then print the final matrix. Not sure how to do this for a

RE: [R] nlme graphics in a loop problem

Use array for your tables and list for plots; e.g., allTable - array(NA, dim=c(numRow, numCol, numIter)) allPlot - vector(mode=list, length=numIter) for (i in 1:numIter) { ... allPlot[[i]] - plot(...) allTable[,,i] - however you generate the table } HTH, Andy From: Jennifer Emond

Re: [R] problem with restore and some .RData

The short answer is to add importFrom(stats, family) to MASS's NAMESPACE and reinstall. Another way is to use library(stats) in your .Rprofile in that directory. The real question is why your .RData is loading namespace MASS. Presumably some object in your workspace has MASS in its

[R] Re: Clustering in R

Thank you for all the responses. I've downloaded TIGR MeV, it seems to do everything I need it to do. The only problem is that none of the algorithms seem to work (K-means, hierarchical) giving errors that just say -2. I think probably the reason for this is that I'm running linux (since there

Re: [R] Resolution of plots

On Thu, 17-Jun-2004 at 08:00AM -0700, Thomas Lumley wrote: | On Wed, 16 Jun 2004, Prof Brian Ripley wrote: | | You will have to tell us more. Exporting how: to what format using what | device and what exact command on what operating system? | | The only device I know of that even knows

Re: [R] Resolution of plots

On Fri, 18 Jun 2004, Patrick Connolly wrote: On Thu, 17-Jun-2004 at 08:00AM -0700, Thomas Lumley wrote: | On Wed, 16 Jun 2004, Prof Brian Ripley wrote: | | You will have to tell us more. Exporting how: to what format using what | device and what exact command on what operating system?

[R] Since package.contents is deprecated, what is replacement function?

I did a ?package.contents in version 1.9.0. The help file says this is deprecated, and says to see also Deprecated and Defunct, but what is the new function that replaces packge.contents? The function still seems to work, but it's just tagged as deprecated. efg [[alternative

[R] [TIP] Apropos resolution of plots

If you need to make JPEG plots for the web or suchlike, here's a method for poor mans antialiasing that seems to turn out rather nice: [Requires ghostscript, netpbm] In R (lifted out of example(contour)) bitmap(out.ppm,ppmraw,res=4*72, pointsize=12) data(volcano) rx - range(x -

Re: [R] Manova question.

On Jun 15, 2004, at 10:11 AM, Douglas Bates wrote: Jens Schumacher wrote: knussear wrote: Hi list, I'm attempting to re-create a Repeated Measures Compositional Analysis as described in the work by Aebischer et. al. (Ecology. 1993. 74(5): 1313-1325). In this paper they describe transitions of

Re: [R] Is there an easy way to generate linearly independent vectors

On 06/17/04 19:04, Fred wrote: Dear R-listers: I am trying to test an algorithm on a set of linearly independent vectors {x1,x2,...,xn}. Well, here's an idea, for 10 vectors of length 10, as columns of a matrix m1. The 11th seems to be needed. m1 - matrix(rnorm(110),10,11) for (i in 2:11) {

RE: [R] Is there an easy way to generate linearly independent vec tors

I believe eigen(), svd() and qr() can all do it. Andy From: Jonathan Baron On 06/17/04 19:04, Fred wrote: Dear R-listers: I am trying to test an algorithm on a set of linearly independent vectors {x1,x2,...,xn}. Well, here's an idea, for 10 vectors of length 10, as columns of a

Re: [R] can't get text to appear over individual panels in multi-panel plot

On Thursday 17 June 2004 19:37, Patrick Bennett wrote: I'm trying to learn how to create Trellis multi-panel plots, but I'm having some trouble reproducing the graphs shown in Venables Ripley (2002) (e.g., Figs 4.14 4.15). Actually, everything looks fine except for the fact that I can't see

Re: [R] Is there an easy way to generate linearly independent vec tors

I want to get linearly independent vectors, not orthogonal ones. The functions eigen, svd, I think it may provide orthogonal vectors which are not what I expect. - Original Message - From: Liaw, Andy [EMAIL PROTECTED] To: 'Jonathan Baron' [EMAIL PROTECTED]; Fred [EMAIL PROTECTED] Cc:

Re: [R] can't get text to appear over individual panels in multi-panel plot

I neglected to say that I am using the R-Aqua interface and the MASS, grid, lattice packages. Here is one specific example where I'm having trouble. After loading the crabs data set, I create the figure with the following code (which is taken from MASS):

Re: [R] can't get text to appear over individual panels in multi-panel plot

On Thursday 17 June 2004 22:24, Patrick Bennett wrote: I neglected to say that I am using the R-Aqua interface and the MASS, grid, lattice packages. Here is one specific example where I'm having trouble. After loading the crabs data set, I create the figure with the following code (which

[R] msm

Hello, I'm writing about msm. It may be that consistent users of Markov models have a good idea as to what constitutes workable data for a model. I think of general rules, in basic statistical studies where n is limited to exclude fairly precise figures in the lower range. On the other

Re: [R] can't get text to appear over individual panels in multi-panel plot

On Thursday 17 June 2004 22:57, Patrick Bennett wrote: yes, i can reproduce that same graph when i print to the pdf-device. but the panel titles do not appear when I print to the Quartz-device. Hmm. I won't be able to help you then, let's hope someone else can. Deepayan On Jun 17, 2004, at