AW: RE: [R] Mass: lda and collinear variables

On one dataset I have 220 cases of physical data (the variables represent points on a curve). The Original data set holds about 1800 variables but I reduce this to a power of 2 (32, 64, ...) with an approximation method so that the representation of one curve (one observation) with say 32 points

Re: AW: RE: [R] Mass: lda and collinear variables

On Mon, 24 Feb 2003, Till Baumgaertel wrote: On one dataset I have 220 cases of physical data (the variables represent points on a curve). The Original data set holds about 1800 variables but I reduce this to a power of 2 (32, 64, ...) with an approximation method so that the representation

[R] error-handling question

hi, I have a question about error-handling. Say I want to do a simulation 100 times, each experiment involves inverting a matrix. Occasionally the matrix is close to singularity, R gives an error message and the program stops right there. Is there any way I can prevent the error message,

RE: [R] error-handling question

Hi, have a look at ?try. Cheers, Winfried On 24-Feb-03 ximing wu wrote: hi, I have a question about error-handling. Say I want to do a simulation 100 times, each experiment involves inverting a matrix. Occasionally the matrix is close to singularity, R gives an error message and the

Re: [R] error-handling question

look at ?try url:www.econ.uiuc.edu Roger Koenker Dept. of Economics UCL, email [EMAIL PROTECTED] Department of Economics Drayton House, vox:217-333-4558University of Illinois 30 Gorden St, fax:217-244-6678Champaign, IL 61820

Re: [R] error-handling question

On Mon, Feb 24, 2003 at 01:50:55AM -0800, ximing wu wrote: hi, I have a question about error-handling. Say I want to do a simulation 100 times, each experiment involves inverting a matrix. Occasionally the matrix is close to singularity, R gives an error message and the program stops

[R] Re: GARCH with t-innovations

Subject: [R] GARCH with t-innovations Date: 21 Feb 2003 14:07:44 +0100 From: Gorazd Brumen [EMAIL PROTECTED] To: [EMAIL PROTECTED] Dear all, Can garch function fit also t-innovations or only Gaussian innovations? -- With kind regards -- Lepo pozdravljeni -- Grüße (Grüezi) --

[R] Rcgi ?

Does anyone know if Rcgiis still available, and if so, where. The link from CRAN http://stats.mth.uea.ac.uk/Rcgi/appears to be broken. cheers Bob .. R.D.Kinley Statistician Eli Lilly co. Liverpool,UK +44 151 448 6347

RE: [R] Subpopulations in Complex Surveys

For svymean, can't you just pass the subpopulation into the design argument? svymean(~crc10yr, design=nhis.design[nhis.design$variables$age=50,], na.rm=TRUE) crc10yr 0.3461349 attr(,var) [,1] [1,] 2.903020e-05 svyglm(crc10yr~I(age=50)+0, design=nhis.design) Stratified 1 - level

[R] gap statistic

Hi, Just one question : Has the gap statistic of R.Tibshirani been implemented in R ? I found nothing on the R website... Thanks in advance Anne-Laure Boulesteix __ [EMAIL PROTECTED] mailing list http://www.stat.math.ethz.ch/mailman/listinfo/r-help

RE: [R] Subpopulations in Complex Surveys

On Mon, 24 Feb 2003, Thompson, Trevor wrote: For svymean, can't you just pass the subpopulation into the design argument? Yes, but you get the wrong standard errors (very slightly) in stratified designs. If you compare the subsetting approach to the regression approach you will find

RE: [R] Problem Writeing a pipe using R (stdin is consumed)

The named pipe idea works for me: basically, $ mknod to_R p $ echo test this out to_R $ R CMD BATCH rtest.R out where rtest.R is con - fifo(to_R, open=r) print(readlines(con)) Reid Huntsinger -Original Message- From: Alberto Gobbi [mailto:[EMAIL PROTECTED] Sent: Friday, February

Re: [R] nlm constrained minimization

Dear all, By the old emails I was told that function constrOptim() is in package r-devel. But where I can find this package? Thanks a lot. Huan Only one question: Can you constrain the minimization in the nlm procedure? No, but you can in optim() and in constrOptim() (the latter in the

[R] convert Splus mapproject() in R

Hello everyone, I was wondering if anyone knows how to convert the Splus mapproject( ) function for use in R, I would greatly appreciate your help. I am using the following system, R Version 1.6.1 (2002-11-01) for SunOS 5.5 and Splus Version 5.1 Release 1 for Sun SPARC, SunOS 5.5. Thanks,

RE: [R] convert Splus mapproject() in R

That function does not seem to exist in Splus 6.1.2 release 2 for Linux. Could it be that it's written by someone locally, or it was in some add-on module? Andy -Original Message- From: Huiqin Yang [mailto:[EMAIL PROTECTED] Sent: Monday, February 24, 2003 10:31 AM To: r-help

Re: [R] convert Splus mapproject() in R

Liaw, Andy wrote: That function does not seem to exist in Splus 6.1.2 release 2 for Linux. Could it be that it's written by someone locally, or it was in some add-on module? Its part of the 'maps' library. Ray Brownrigg has ported it to R, but looking at the DESCRIPTION file tells me this:

[R] Legend in plot: symbol for mean and standard deviation

Dear list, I am facing the following problem with the legend of a plot that display the mean and variance of a measurement y as a function of x, the mean being represented by a dot and the variance by a vertical line. My problem is that I am unable to display the symbol (dot + vertical line) in

[R] segmented regression

Hello all, I'm trying to find out how to perform a 'segmented regression'. I have some data and the 'classical' model used is: y(t) = a + bx(t) + cx(t)^2 + u(t) for x(t) x(0) y(t) = a + bx(0) + cx(0)^2 + d(x(t) - x(0)) + u(t) for x(t) x(0) and u(t) = rho.u(t-1) + eps (It's a model using an

Re: [R] calculating mean direction (CircStats)

Cordula == Cordula Becker [EMAIL PROTECTED] on Fri, 21 Feb 2003 20:39:00 +0100 writes: Cordula Hi, I've currently to work with some circular Cordula data. Unfortunately I'm not very familiar with Cordula circular statistics and would really appreciate if Cordula I could get

[R] bwplot stats question

Hi List, Just wondering where the documentation exists for the statistics which makeup the bwplot. I'm guessing that if R is like similar products that the graph is constructed as The median is the filled circle. The box surrounding the filled circle depicts the 25th and 75th quartile. The

RE: [R] printing decimal numbers

-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Gerds Sent: Monday, February 24, 2003 11:01 AM To: [EMAIL PROTECTED] Subject: [R] printing decimal numbers hi, this is a very basic question -- sorry for posing it: how can i force R to print

RE: [R] bwplot stats question

bwplot() calls boxplot.stats() for the statistics, so look up ?boxplot.stats. Andy -Original Message- From: Ken Nussear [mailto:[EMAIL PROTECTED] Sent: Monday, February 24, 2003 11:44 AM To: [EMAIL PROTECTED] Subject: [R] bwplot stats question Hi List, Just wondering where

Re: [R] bwplot stats question

The documentation is 1) the code 2) ?boxplot 3) the standard references on EDA. I'd have to look up which exactly, but my guess would be to look at Velleman Hoaglin (1981) ABC of EDA. Hoaglin, Mosteller, Tukey (1983) Understanding Robust and Exploratory Data Analysis. Some better references

Re: [R] bwplot stats question

On Mon, Feb 24, 2003 at 08:43:56AM -0800, Ken Nussear wrote: ... Just wondering where the documentation exists for the statistics which makeup the bwplot. It goes: bwplot calls panel.bwplot, which calls boxplot.stats help(boxplot.stats) gives you all you need to know, and more. The

Re: [R] faraway tutorial: cryptic command to newbie

That's an unfair criticism. That discussion was never intended as a recommendation for how to compute a regression. Of course, SVD or QR decompositions are the preferred method but many newbies don't want to digest all that right from the start. These are just obscure details to the beginner.

Re: [R] Legend in plot: symbol for mean and standard deviation

Marco Kienzle wrote: Dear list, I am facing the following problem with the legend of a plot that display the mean and variance of a measurement y as a function of x, the mean being represented by a dot and the variance by a vertical line. At least for me the latter does not appear to be

Re: [R] exact range of axes in plots

Hope this helps. Jerome plot(0:1,0:1) par()$usr [1] -0.04 1.04 -0.04 1.04 On Monday 24 February 2003 11:06, Jason Bond wrote: Hello. I was wondering how one can find the exact values of the range of an axis within a plot. In xlispstat it was (send plot :range 0). Thanks much,

Re: [R] exact range of axes in plots

Jason Bond wrote: Hello. I was wondering how one can find the exact values of the range of an axis within a plot. In xlispstat it was (send plot :range 0). Thanks much, After you have created the plot try par(usr). This gives 4 values: xmin, xmax, ymin, ymax. plot(1:10, 1:10) par(usr) [1]

Re: [R] exact range of axes in plots

Hi, - Original Message - From: Jason Bond [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Tuesday, February 25, 2003 8:06 AM Subject: [R] exact range of axes in plots Hello. I was wondering how one can find the exact values of the range of an axis within a plot. In xlispstat it was

RE: [R] exact range of axes in plots

-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jason Bond Sent: Monday, February 24, 2003 1:06 PM To: [EMAIL PROTECTED] Subject: [R] exact range of axes in plots Hello. I was wondering how one can find the exact values of the range of an axis within

[R] creating a plot legend

Thanks so much to all for their help. I was wondering if there was an easy way to create legends in R plots. For example, if I am plotting lines for subjects in a longitudinal design of the outcome by time for two different treatments, where the individual for treatment 1 are plotted in one

Re: [R] creating a plot legend

On Mon, 24 Feb 2003, Jason Bond wrote: Thanks so much to all for their help. I was wondering if there was an easy way to create legends in R plots. For example, if I am plotting lines for subjects in a longitudinal design of the outcome by time for two different treatments, where the

[R] Test suites

I have a collection of functions, class definitions and methods which I would like to test systematically for their correctness after changes to their code, and also after major R revisions. I believe that the correct term for these systematic tests (as opposed to more informal tests) is a 'test

Re: [R] Test suites

all.equal described in Venables and Ripley (MASS, 4th ed., p. 49, and S Programming, p. 21) is useful. (The index entry in S Programming is off by 1 page.) Hope this helps. Spencer Graves Giles Heywood wrote: I have a collection of functions, class definitions and methods which I would like to

RE: [R] Forcing plots to adjust-to-data

-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jason Bond Sent: Monday, February 24, 2003 5:12 PM To: [EMAIL PROTECTED] Subject: [R] Forcing plots to adjust-to-data Hello. In messing around with some of the graphical parameters using the par()

Re: [R] question on - build R as a shared library

On Mon, Feb 24, 2003 at 08:34:24PM -0800, Bai Yan wrote: I used '--enable-R-shlib' as the argument for 'configure'. After the installation, I want to check whether R is built as a shared library since I need to assure this for installing other packages like SJava. Could anybody tell me how to

Re: [R] question on - build R as a shared library

Hi Dirk, thank you for the prompt reply. I found libR.so. But how can I know whether it is build as shared or not. On Mon, 24 Feb 2003, Dirk Eddelbuettel wrote: On Mon, Feb 24, 2003 at 08:34:24PM -0800, Bai Yan wrote: I used '--enable-R-shlib' as the argument for 'configure'. After the