[R] Query regarding modelling using R
Hi, I am working on project to build a model for analysing Mortgage-backed securities. I am doing this as a part of my summer project. After much thought I have decided on using R for doing this modelling. One of the steps in this model is making an interest rate model to generate interest rate paths in time. My question is.. Is any package available to do the same in R? or I will have to build a function myself from scratch? Any kind of help will be appreciated. Thank You David Flash. [[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] Crosstabs
Dear Users, I'm a complete novice to R. I need to do a crosstabs in R, but my data is almost completely alphanumeric (with some variables scaled). The Table routine does not seem to accept alphanumeric data. What should I do? Do I need to recode it? How should I do that? Thanks in advance, Wilfred __ 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] RMySQL Suse 10.0 installing problems
On Thu, 6 Jul 2006, H. Paul Benton wrote: Thanks again everyone, I have a problem trying to install RMySQL on Suse 10.0. I have used both the install.packages() and the 'R CMD INSTALL RMySQL... neither worked. I get the error message telling me that it cannot find mysql. So I have set the flags using export PKG_CPPFLAGS=-I/usr/bin/ export PKG_LIBS=-L/usr/lib/mysql Those cannot be right: you don't want and -I/usr/bin is not a plausible include path. RPM-based systems normally need export PKG_CPPFLAGS=-I/usr/include/mysql export PKG_LIBS=-I/usr/lib/mysql and if you have /usr/lib/mysql, try /usr/lib/mysql/mysql_config -cflags to be sure about the first. After that I run R CMD INSTALL RMySQL... it starts to check and it cannot find mysql_init, mysql.h then it exits after gcc(ing) with an error code of make:*** [RS-DBI.o] Error 1 I also tried the other way that the help suggested of putting the dir in the R CMD INSTALL command but that didn't work either giving me the same help menu. I am on opensuse 10.0 x86, and yes I su(ed). Cheers, Research Technician Mass Spectrometry o The / o Scripps \ o Research / o Institute __ 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 -- 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] engineering notation format
Both format and formatC (and sprintf) use C facilities for scientific format. As far as I know C has no facilities for your desired format, so you would need to write your own C code and interface to R. On Thu, 6 Jul 2006, Walker, Sam wrote: Hi, How can I format numbers to engineering notation? That is, like scientific but where the exponent is always a multiple of three. Some examples: 1635 000 000 = 1.635E9 163 500 000 = 163.5E6 0.000 000 000 135 = 135E-9 I've tried format and formatC but I couldn't get them to work they I want. Any help is greatly appreciated. Thanks, Sam __ 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 -- 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] KPSS test
Hello Sachin, a sequential testing procedure is described in the useR! book: @Book{, title = {Analysis of Integrated and Cointegrated Time Series with R}, author = {B. Pfaff}, publisher = {Springer}, edition = {First}, address = {New York}, year = {2006}, note = {ISBN 0-387-27960-1}, } Best, Bernhard Dr. Bernhard Pfaff Global Structured Products Group (Europe) Invesco Asset Management Deutschland GmbH Bleichstrasse 60-62 D-60313 Frankfurt am Main Tel: +49(0)69 29807 230 Fax: +49(0)69 29807 178 Email: [EMAIL PROTECTED] -Ursprüngliche Nachricht- Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Im Auftrag von Sachin J Gesendet: Donnerstag, 6. Juli 2006 21:49 An: [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Betreff: Re: [R] KPSS test Hi Mark, Thanx for the help. I will verify my results with PP and DF test. Also as suggested I will take a look at the references pointed out. One small doubt: How do I decide what terms ( trend, constant, seasonality ) to include while using these stationarity tests. Any references would be of great help. Thanx, Sachin [EMAIL PROTECTED] wrote: From: Date: Thu Jul 06 14:17:25 CDT 2006 To: Sachin J Subject: Re: [R] KPSS test sachin : i think your interpretations are right given the data but kpss is quite a different test than the usual tests because it assumes that the null is stationarity while dickey fuller ( DF ) and phillips perron ( PP ) ) assume that the null is a unit root. therefore, you should check whetheer the conclusions you get from kpss are consistent with what you would get from DF or PP. the results often are not consistent. also, DF depends on what terms ( trend, constant ) you used in your estimation of the model. i'm not sure if kpss does also. people generally report Dickey fuller results but they are a little biased towards acepting unit root ( lower power ) so maybe that's why you are using KPSS ? Eric Zivot has a nice explanation of a lot of the of the stationarity tests in his S+Finmetrics book. testing for cyclical variation is pretty complex because that's basically the same as testing for seasonality. check ord's or ender's book for relatively simple ways of doing that. From: Sachin J Date: Thu Jul 06 14:17:25 CDT 2006 To: R-help@stat.math.ethz.ch Subject: [R] KPSS test Hi, Am I interpreting the results properly? Are my conclusions correct? KPSS.test(df) KPSS test Null hypotheses: Level stationarity and stationarity around a linear trend. Alternative hypothesis: Unit root. Statistic for the null hypothesis of level stationarity: 1.089 Critical values: 0.10 0.05 0.025 0.01 0.347 0.463 0.574 0.739 Statistic for the null hypothesis of trend stationarity: 0.13 Critical values: 0.10 0.05 0.025 0.01 0.119 0.146 0.176 0.216 Lag truncation parameter: 1 CONCLUSION: Reject Ho at 0.05 sig level - Level Stationary Fail to reject Ho at 0.05 sig level - Trend Stationary kpss.test(df,null = c(Trend)) KPSS Test for Trend Stationarity data: tsdata[, 6] KPSS Trend = 0.1298, Truncation lag parameter = 1, p-value = 0.07999 CONCLUSION: Fail to reject Ho - Trend Stationary as p-value sig. level (0.05) kpss.test(df,null = c(Level)) KPSS Test for Level Stationarity data: tsdata[, 6] KPSS Level = 1.0891, Truncation lag parameter = 1, p-value = 0.01 Warning message: p-value smaller than printed p-value in: kpss.test(tsdata[, 6], null = c(Level)) CONCLUSION: Reject Ho - Level Stationary as p-value sig. level (0.05) Following is my data set structure(c(11.08, 7.08, 7.08, 6.08, 6.08, 6.08, 23.08, 32.08, 8.08, 11.08, 6.08, 13.08, 13.83, 16.83, 19.83, 8.83, 20.83, 17.83, 9.83, 20.83, 10.83, 12.83, 15.83, 11.83), .Tsp = c(2004, 2005.917, 12), class = ts) Also how do I test this time series for cyclical varitions? Thanks in advance. Sachin - [[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 - [[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 * Confidentiality Note: The information contained in this mess...{{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
[R] Polynomial kernel in SVM in e1071 package
Dear list, In some places (for example, http://en.wikipedia.org/wiki/Support_vector_machine) , the polynomail kernel in SVM is written as (u'*v + 1)^d, while in the document of svm() in e1071 package, the polynomial kernel is written as (gamma*u'*v + coef0)^d. I am a little confused here: When doing parameter optimization (grid search or so) for polynomial kernel, does it need to tune four parameters, gamma, coef0, C and degree, or just two of them, C and degree (and fixing gamma to 1 and coef0 = 1)? Thanks, Wuming __ 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 crash with 'library(Matrix); as(x, dgCMatrix)' ...
JohnT == Thaden, John J [EMAIL PROTECTED] on Thu, 6 Jul 2006 12:29:42 -0500 writes: JohnT Martin Maechler replied to my query Warning while subsetting...: MartinM JohnT == Thaden, John J [EMAIL PROTECTED] MartinM on Thu, 6 Jul 2006 00:02:10 -0500 writes: JohnT ... JohnT # While subsetting x, I was surprised to get this warning: JohnT y-x[1:300,] JohnT Warning message: JohnT number of items to replace is not a multiple of JohnT replacement length MartinM and later JohnT Sorry, I omitted background information: JohnT R version: 2.3.0 JohnT OS: Windows XP JohnT CPU: Pentium III, JohnT RAM: 768 MB MartinM You omitted the most pertinent information: The MartinM version of 'Matrix' you are using. MartinM The latest released version of Matrix does MartinM *not* show the behavior you mentioned. {So I have MartinM now spent 20 minutes just because you did not MartinM update 'Matrix'..} JohnT The Matrix package was version 0.995-10, now is 0.995-11. JohnT The R base was version 2.3.0, now is 2.3.1. JohnT Subsetting 'y - x[1:300,]' now works. Please accept my apology. JohnT Also, what command-line memory settings might prevent JohnT R from crashing while using the Matrix package to JohnT convert my 600 X 4482 dgTMatrix to the dgCMatrix class JohnT or to an expanded Matrix, via the as() function? I can JohnT do this with half of the matrix, 300 x 4482. MartinM It's hard to believe that you get a crash MartinM when coercing to 'dgC' -- but of course this MartinM really depends how much memory you have already MartinM goggled up by other large objects in your R MartinM workspace, or by other applications running at MartinM the same time in Windows. Coercing to a full MartinM matrix will of course require 8 * 601 * 4482 = MartinM 21549456 extra bytes just for the numbers. MartinM That's only 21.5 Megabytes, so I wonder.. MartinM MartinM I have never seen R crashes from using 'Matrix', (actually that's not even true; at some point in time we had a bug in 'Matrix' which lead to spurious segmentation faults) MartinM but then I work with an operating system, not MartinM with M$ Windows. MartinM MartinM Maybe you meant you got an error message MartinM ... memory allocation ..? JohnT Testing again, I closed all applications; disabled antivirus; JohnT opened RGui; removed all R objects but 'x' (a 600x4482 dgTMatrix); JohnT opened WinXP's 'Task Manager'; saw only Rgui under JohnT 'Applications'; saw processes using a total of 287 MB of memory JohnT under 'Processes'; closed 'Task Manager'; and typed R commands: # Steps leading to an R crash... ls() JohnT [1] x str(x) JohnT Formal class 'dgTMatrix' [package Matrix] with 6 slots JohnT ..@ i : int [1:923636] 1 2 3 4 5 6 7 8 9 10 ... JohnT ..@ j : int [1:923636] 1 1 1 1 1 1 1 1 1 1 ... JohnT ..@ Dim : int [1:2] 600 4482 JohnT ..@ Dimnames:List of 2 JohnT .. ..$ : chr [1:601] 50 51 52 53 ... JohnT .. ..$ : chr [1:4482] 1 2 3 4 ... JohnT ..@ x : num [1:923636] 50.2 51.2 52.2 53.2 54.2 ... JohnT ..@ factors : list() gc() JohnT used (Mb) gc trigger (Mb) max used (Mb) JohnT Ncells 183529 5.0 407500 10.9 35 9.4 JohnT Vcells 1928101 14.82286173 17.5 1928652 14.8 library(Matrix) JohnT Loading required package: lattice gc() JohnT used (Mb) gc trigger (Mb) max used (Mb) JohnT Ncells 627772 16.81073225 28.7 1073225 28.7 JohnT Vcells 2165773 16.63345184 25.6 2332013 17.8 search() JohnT [1] .GlobalEnv package:Matrix package:lattice JohnT [4] package:methods package:stats package:graphics JohnT [7] package:grDevices package:utils package:datasets JohnT [10] Autoloads package:base #Now the line that causes crashes... y - as(x,dgCMatrix) JohnT After ~10 seconds, R blinks off and a WinXP dialog appears: JohnT R for Windows GUI front-end has encountered JohnT a problem and needs to close. We are sorry JohnT for the inconvenienceError signature: JohnT AppName: rgui.exe AppVer: 2.31.38247.0 JohnT ModName: matrix.dll Offset: ff31 JohnT Report error? Thanks a lot, John, for the more detailed report. I do wonder how it happens, since the memory allocation is not really big. E.g., I can easily solve ``your'' (well, a simulated version of it) problem on a machine with only 512 MB RAM: library(Matrix) ## MM: construct a matrix *as* John's : d - as.integer(c(600,4482)) n0 - 923636 set.seed(1) M - new(dgTMatrix, Dim = d, i = sort(sample(0:(d[1]-1), size = n0, replace = TRUE)), j = sample(0:(d[2]-1), size = n0, replace = TRUE), x = round(rnorm(n0, m = 50, sd = 10), 1))
[R] how to name a variable?
Dear friends, The s in the following argument don't have a variable name, how should i give it a name? s-data.frame(seq(1,6,by=2)) s seq.1..6..by...2. 1 1 2 3 3 5 thanks very much! -- Kind Regards, Zhi Jie,Zhang ,PHD Department of Epidemiology School of Public Health Fudan University Tel:86-21-54237149 [[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] engineering notation format
Hi, try this: formatEng - function(x) { s-as.numeric(strsplit(format(x, scientific=T),e)[[1]]) return(paste(s[1]*10^(s[2]%%3),as.integer(s[2]-(s[2]%%3)),sep=e)) } Some examples: 1635 000 000 = 1.635E9 163 500 000 = 163.5E6 0.000 000 000 135 != 135E-9 0.000 000 000 135 = 125E-12 ? Hans ** Hans-Joerg Bibiko Max Planck Institute for Evolutionary Anthropology Department of Linguistics Deutscher Platz 6 phone: +49 (0) 341 3550 341 D-04103 Leipzig fax: +49 (0) 341 3550 333 Germany 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] how to name a variable?
s-data.frame(var.name=seq(1,6,by=2)) s var.name 11 23 35 2006/7/7, zhijie zhang [EMAIL PROTECTED]: Dear friends, The s in the following argument don't have a variable name, how should i give it a name? s-data.frame(seq(1,6,by=2)) s seq.1..6..by...2. 1 1 2 3 3 5 thanks very much! -- Kind Regards, Zhi Jie,Zhang ,PHD Department of Epidemiology School of Public Health Fudan University Tel:86-21-54237149 [[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 -- 黄荣贵 Department of Sociology Fudan University __ 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] Start Model for POLYCLASS
I have not seen a reply, so I will offer a few suggestions, even though I've never used the 'polspline' package. I scanned several of the help pages and looked in ~\library\polspline' where R is installed on my hard drive and found no further documentation, and I found nothing new from RSiteSearch(polyclass). Googling for 'polyclass' led me to http://bear.fhcrc.org/~clk/soft.html;, the home page for Charles Kooperberg, the author and maintainer. If it were my problem, I might try writing him directly at the email address given in help(package=polyclass); I'm including him as a cc on this reply. I spent time looking at this, because 'polyclass', 'polymars' and 'polspline' seem potentially related related to one of my secondary interests. Unfortunately, I couldn't find sufficient documentation to allow me to proceed with the time I felt I could afford to invest in this right now. If it were my problem, before I wrote another email about this, I'd first list the function 'polyclass', make a local copy, then work through an example line by line using 'debug(polyclass)'. This 'debug' facility is remarkably powerful and easy to use. I've solved many problems like this using 'debug' in this way. If that failed to provide the necessary enlightenment, I'd submit another post including a self-contained example based on a modification of the 'iris' example featured in the 'polyclass' help page. Your example is NOT self contained, so I would NOT use that. Using the 'iris' example would make it much easier to explain clearly what you want. It also makes it much easier for someone like me to experiment with alternatives and describe what I did in terms of a tested example. Hope this helps. p.s. I suggest you also review the posting guide! www.R-project.org/posting-guide.html. Xiaogang Su wrote: Dear all, I have a question on how to set up the starting model in POLYCLASS and make sure the terms in the starting model retained in the final POLYCLASS model. In the function POLYMARS, this can be done using the STARTMODEL option. See below for example, I started with model y= b0 + b1*X1 + b2*X2 + b3*X4 + b4*X5 + b5*X2*X5 + e m00 - matrix(c( 1, NA, 0, NA, 1, 2, NA, 0, NA, 1, 4, NA, 0, NA, 1, 5, NA, 0, NA, 1, 2, NA, 5, NA, 1),nrow = 5, ncol=5, byrow=TRUE); m2 - polymars(response=PID2$y, predictors=PID2[,1:7], startmodel=m00) summary(m2) But I could not figure out how this works for POLYCLASS. There is an option FIT in POLYCLASS, which needs to be a POLYCLASS object though. Any suggestion or information is greatly appreciated. Sincerely, Xiaogang Su Xiaogang Su, Assistant Professor Department of Statistics and Actuarial Science University of Central Florida Orlando, FL 32816 (407) 823-2940 [O] [EMAIL PROTECTED] http://pegasus.cc.ucf.edu/~xsu/ __ 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] Query regarding modelling using R
flash johny flash.johny at gmail.com writes: Hi, I am working on project to build a model for analysing Mortgage-backed securities. I am doing this as a part of my summer project. Try http://cran.r-project.org/src/contrib/Views/ Dieter __ 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] Crosstabs
Bi-Info (http://members.home.nl/bi-info) wrote: Dear Users, I'm a complete novice to R. I need to do a crosstabs in R, but my data is almost completely alphanumeric (with some variables scaled). The Table routine does not seem to accept alphanumeric data. What should I do? Do I need to recode it? How should I do that? Works for me: table(c(a, a, b), c(a, c, c)) Uwe Ligges Thanks in advance, Wilfred __ 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] Polynomial kernel in SVM in e1071 package
Wuming Gong wrote: Dear list, In some places (for example, http://en.wikipedia.org/wiki/Support_vector_machine) , the polynomail kernel in SVM is written as (u'*v + 1)^d, while in the document of svm() in e1071 package, the polynomial kernel is written as (gamma*u'*v + coef0)^d. I am a little confused here: When doing parameter optimization (grid search or so) for polynomial kernel, does it need to tune four parameters, gamma, coef0, C and degree, or just two of them, C and degree (and fixing gamma to 1 and coef0 = 1)? All of them, unless you know reasonable values, but that's very improbable. Uwe Ligges Thanks, Wuming __ 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] Diverging results with SPSS
Dear List, I apologize in advance if this is silly. I tried to replicate an analysis I did previously in SPSS using R, and was surprised to find different results. So my question is: shouldn't the following SPSS syntax REGRESSION DEPENDENT INC89 /METHOD=ENTER hiedyrs experien SE93rec. Yeld the same results of the following R command modelB-lm(INC89~HIEDYRS+EXPERIEN+SE93REC) I assume the is some difference in some default options.Or maybe it was a problem when the data was imported. After using the read.spss in the foreign package, I got the following warning message: Unrecognized record type 7, subtype 16 encountered in system file Thanks in advance for the help and apologies for the trouble, Celso [[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] Diverging results with SPSS
Celso Barros [EMAIL PROTECTED] writes: Dear List, I apologize in advance if this is silly. I tried to replicate an analysis I did previously in SPSS using R, and was surprised to find different results. So my question is: shouldn't the following SPSS syntax REGRESSION DEPENDENT INC89 /METHOD=ENTER hiedyrs experien SE93rec. Yeld the same results of the following R command modelB-lm(INC89~HIEDYRS+EXPERIEN+SE93REC) Possibly, if all variables are quantitative. What are the differences that you see? I assume the is some difference in some default options.Or maybe it was a problem when the data was imported. After using the read.spss in the foreign package, I got the following warning message: Unrecognized record type 7, subtype 16 encountered in system file You might check whether your variables have the same summary statistics in both systems. Thanks in advance for the help and apologies for the trouble, Celso [[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 -- 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] Diverging results with SPSS
If results are *very* different, I would suspect that R treated experien as a factor (when you use read.spss the default is to use value labels, which makes R treat such variables as factors -- set use.value.labels = F). Another explanation could be missing data in SPSS (not sysmis, but coded with, say -99) that are not recognized by R. Florian Von: [EMAIL PROTECTED] im Auftrag von Celso Barros Gesendet: Fr 07.07.2006 07:13 An: r-help@stat.math.ethz.ch Betreff: [R] Diverging results with SPSS Dear List, I apologize in advance if this is silly. I tried to replicate an analysis I did previously in SPSS using R, and was surprised to find different results. So my question is: shouldn't the following SPSS syntax REGRESSION DEPENDENT INC89 /METHOD=ENTER hiedyrs experien SE93rec. Yeld the same results of the following R command modelB-lm(INC89~HIEDYRS+EXPERIEN+SE93REC) I assume the is some difference in some default options.Or maybe it was a problem when the data was imported. After using the read.spss in the foreign package, I got the following warning message: Unrecognized record type 7, subtype 16 encountered in system file Thanks in advance for the help and apologies for the trouble, Celso [[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 [[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] R-sig-foo searching, was : High breakdown/efficiency statistics -- was RE: Rosner's test
Le 23.06.2006 10:42, Martin Maechler a écrit : {BTW: Did you know that to *search* mailing list archives of such R-SIG-foo mailing lists, you can use google very efficiently by prepending the mailing list name and 'site:stat.ethz.ch'? e.g., use google search on R-SIG-robust site:stat.ethz.ch lmrob } Great idea, with a little bonus for the gnome mini-commander lovers (the other folks can just ignore this), a new macro : expression : |^rs-(.*):(.*)$| command : gnome-open http://www.google.fr/search?q=R-SIG-\1 \2 site:stat.ethz.ch Now you can type rs-robust:lmrob on your gnome mini commander applet Cheers, Romain -- visit the R Graph Gallery : http://addictedtor.free.fr/graphiques mixmod 1.7 is released : http://www-math.univ-fcomte.fr/mixmod/index.php +---+ | Romain FRANCOIS - http://francoisromain.free.fr | | Doctorant INRIA Futurs / EDF | +---+ __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] convert ms() to optim()
How to convert the following ms() in Splus to Optim in R? The Calc function is also attached. ms(~ Calc(a.init, B, v, off, d, P.a, lambda.a, P.y, lambda.y, 10^(-8), FALSE, 20, TRUE)$Bic, start = list(lambda.a = 0.5, lambda.y = 240), control = list(maxiter = 10, tol = 0.1)) Calc - function(A.INIT., X., V., OFF., D., P1., LAMBDA1., P2., LAMBDA2., TOL., MONITOR., MAX.ITER., TRACE.){ lambda1 - abs(LAMBDA1.) lambda2 - abs(LAMBDA2.) P - lambda1 * P1. + lambda2 * P2. a - Estimate(A.INIT., X., V., OFF., D., P, TOL., MONITOR., MAX.ITER.) Ita - OFF. + X. %*% a Mu - c(exp(Ita)) Wt - Mu * V. Bt.W.B - t(X.) %*% (Wt * X.) BtWBplusP - Bt.W.B + P Rhs - Bt.W.B %*% a + t(X.) %*% (V. * (D. - Mu)) a - solve(BtWBplusP, Rhs) Tr - sum(diag(solve(BtWBplusP, Bt.W.B))) y.init - D. y.init[D.==0] - 10^(-4) Dev - 2*sum( V. * D.*log(y.init/Mu) ) Bic - Dev + log(sum(V.)) * Tr Hazard - Ita - OFF. if (TRACE. == TRUE) cat(lambda1, lambda2, Bic, \n) return(a, Hazard, Tr, Dev, Bic, BtWBplusP) Thanks, Shelton - [[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] Calling R from Java
Madam, I've read your answer : /// Here are other alternatives that are easy to implement and that you should consider depending on what you want to do: 1- one is to use R in batch mode ie you create a file in your java code with all the R commands and send it to R (via Process pc=Runtime.getRuntime().exec(Rcall); )and then R creates an outputfile with all the results. The matter with this is that there is no interactivity, variables are not kept, you have to parse the outputfile, ... (see the above article (see § FFinterface)) 2- another one is to use the R server developed by Simon Urbanek: http://stats.math.uni-augsburg.de/Rserve/ A java frontend is provided with the server, which allows to communicate easily with R and which sends back results in a java object. So for each connection to the server you have a R environment where all the variable generated are kept until you close the connection and from you java code it is as simple as this example: REXP mean = c.eval(mean+i+-apply(+data+[,data.cl==+(i-1)+],1,mean.na)); This is what I'am using in my application and I am quite satisfied with it. // For the first solution, I have not understood how I get the results from R??? For the second solution, I didn't understant the instruction, i'm beginner in R and Java??? I ve understood that Rserve is a server that I should install server R, R and the JDK for the jav??? Thank you in advance, XENA - [[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] engineering notation format
Clever... Good observation on the last example, mistake on my part. Thanks for all the suggestions. Sam -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Hans-Joerg Bibiko Sent: Friday, July 07, 2006 4:27 AM To: r-help@stat.math.ethz.ch Subject: Re: [R] engineering notation format Hi, try this: formatEng - function(x) { s-as.numeric(strsplit(format(x, scientific=T),e)[[1]]) return(paste(s[1]*10^(s[2]%%3),as.integer(s[2]-(s[2]%%3)),sep=e)) } Some examples: 1635 000 000 = 1.635E9 163 500 000 = 163.5E6 0.000 000 000 135 != 135E-9 0.000 000 000 135 = 125E-12 ? Hans ** Hans-Joerg Bibiko Max Planck Institute for Evolutionary Anthropology Department of Linguistics Deutscher Platz 6 phone: +49 (0) 341 3550 341 D-04103 Leipzig fax: +49 (0) 341 3550 333 Germany 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 __ 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] parametric proportional hazard regression
Dear all, I am trying to find a suitable R-function for parametric proportional hazard regressions. The package survival contains the coxph() function which performs a Cox regression which leaves the base hazard unspecified, i.e. it is a semi-parametric method. The package Design contains the function pphsm() which is good for parametric proportional hazard regressions when the underlying base distribution is weibull or exponential. But what if I need a parametric proportional hazard model with the other usual distributions like 'gaussian', 'logistic', 'lognormal'and 'loglogistic? Thanks a lot for your support! Valentin Dimitrov Statistics and Econometrics University of Saarland __ 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] replace values in data frame
Hi all, I have a three columned list that I have imported into R. The first column is a plot (ex. Plot1), the second is a species name (ex ACERRUB) and the third a numeric value. I want to replace some of the second column names with other names (for example replace ACERRUB with ACERDRU). The original and replacement values are in separate lists (not vectors), but I can't seem to find the right function to perform this. The replace function seems to only want to work with numbers. Any clues? Wade [[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] dotplot (lattice) with panel.segments and groups
Hi, The following produces almost exactly what I needed. The problems are that the 'panel.dotplot' call (commented) generates the error 'Error in NextMethod([) : argument subscripts is missing, with no default'. The other problem is that the colors alternate between the levels of the 'site' variable, rather than 'year'. barley$yield2 - with(barley, yield + 5) dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel=function(x, y, subscripts, yield2, ...) { ## panel.dotplot(x, y, ...) panel.segments(x, as.numeric(y), yield2[subscripts], as.numeric(y), ...) }) R sessionInfo() Version 2.3.1 (2006-06-01) i486-pc-linux-gnu attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: chron gmodels lattice 2.3-3 2.12.0 0.13-8 Can somebody please suggest how to properly make that 'panel.dotplot' call and set the 'col' argument? Thanks in advance. Cheers, -- Seb __ 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] replace values in data frame
On Fri, 2006-07-07 at 11:20 -0400, Wade Wall wrote: Hi all, I have a three columned list that I have imported into R. The first column is a plot (ex. Plot1), the second is a species name (ex ACERRUB) and the third a numeric value. I want to replace some of the second column names with other names (for example replace ACERRUB with ACERDRU). The original and replacement values are in separate lists (not vectors), but I can't seem to find the right function to perform this. The replace function seems to only want to work with numbers. Any clues? Wade Without seeing the code you are using, we can only guess a syntax error of some sort. It works fine using the iris dataset: head(iris) Sepal.Length Sepal.Width Petal.Length Petal.Width Species 1 5.1 3.5 1.4 0.2 setosa 2 4.9 3.0 1.4 0.2 setosa 3 4.7 3.2 1.3 0.2 setosa 4 4.6 3.1 1.5 0.2 setosa 5 5.0 3.6 1.4 0.2 setosa 6 5.4 3.9 1.7 0.4 setosa 7 4.6 3.4 1.4 0.3 setosa 8 5.0 3.4 1.5 0.2 setosa 9 4.4 2.9 1.4 0.2 setosa 10 4.9 3.1 1.5 0.1 setosa iris$Species - replace(iris$Species, iris$Species == setosa, NewValue) head(iris) Sepal.Length Sepal.Width Petal.Length Petal.Width Species 1 5.1 3.5 1.4 0.2 NewValue 2 4.9 3.0 1.4 0.2 NewValue 3 4.7 3.2 1.3 0.2 NewValue 4 4.6 3.1 1.5 0.2 NewValue 5 5.0 3.6 1.4 0.2 NewValue 6 5.4 3.9 1.7 0.4 NewValue 7 4.6 3.4 1.4 0.3 NewValue 8 5.0 3.4 1.5 0.2 NewValue 9 4.4 2.9 1.4 0.2 NewValue 10 4.9 3.1 1.5 0.1 NewValue HTH, Marc Schwartz __ 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] How to change the type of segments ends?
Hi, I am trying to plot odds ratios and the corresponding confidence intervals in horizontal segments. It would be ideal if the confidence interval segment can be drawn with little vertical bars at both ends. I have tried very hard to change the type of ends by using 'lend' arguments, but cannot make it. I even tried 'arrows()', but still failed. Following is the code I use: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x=drug.or,y=yaxis,type='p',pch=17,xlim=c(0,3),axes=FALSE, xlab='Odds Ratio',ylab='',main='Reference Group: A only') axis(1,at=seq(0,3,by=0.5),labels=paste(seq(0,3,by=0.5))) axis(2,at=yaxis,las=2) segments(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,col=4,lend=2) # or try #arrows(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,length=0.1,angle=0,cod e=3,col=4,lend=2) box() = Any comments or suggestions would be greatly appreciated. Jane University of Pittsburgh Pittsburgh, PA 15261 __ 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] graphlets
Naras presented something with similar functionalities using SVG a couple of years ago, and since then the svgdevice package had been made available. I do not know SVG myself, but according to Naras, it's all possible. Best, Andy From: Terry Therneau I have an application where the Splus graphlets() package would work very well, and would like to know if there is any comparable work in R. In response to a similar query on this list about 1.5 years ago the Orca package was mentioned, but it seems to be inactive. For those who don't know, graphlets are an extended java graphics device. The identify function, for instance, allows one to link both a label and a set of html links to each point. I am aware of the Sjava package in Omega, and am exploring that, but would be very interested if there is anything more. Terry Therneau Mayo Clinic __ R-help@stat.math.ethz.ch mailing list https://www.stat.math.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] Diverging results with SPSS
Dear Peter and Florian, I think you have spotted the problem, thank you very much. I checked the frequencies, and R is not understanding that the (-1) value in a binary variable was defined in SPSS as sysmis. This was precisely the variable that showed the larger difference (it's the only binary variable in the model). Thanks a lot. I am still trying to make R understand that, though. If I could avoid defining missing values inside R for each of the variables in the dataset, I would be happy. Does anyone know if there is a command in foreign, for instance, that ensures that things coded as sysmis in SPSS are read as sysmis in R? Thanks again, this has been very, very helpful Celso On 7/7/06, Florian Koller [EMAIL PROTECTED] wrote: If results are *very* different, I would suspect that R treated experien as a factor (when you use read.spss the default is to use value labels, which makes R treat such variables as factors -- set use.value.labels = F). Another explanation could be missing data in SPSS (not sysmis, but coded with, say -99) that are not recognized by R. Florian -- *Von:* [EMAIL PROTECTED] im Auftrag von Celso Barros *Gesendet:* Fr 07.07.2006 07:13 *An:* r-help@stat.math.ethz.ch *Betreff:* [R] Diverging results with SPSS Dear List, I apologize in advance if this is silly. I tried to replicate an analysis I did previously in SPSS using R, and was surprised to find different results. So my question is: shouldn't the following SPSS syntax REGRESSION DEPENDENT INC89 /METHOD=ENTER hiedyrs experien SE93rec. Yeld the same results of the following R command modelB-lm(INC89~HIEDYRS+EXPERIEN+SE93REC) I assume the is some difference in some default options.Or maybe it was a problem when the data was imported. After using the read.spss in the foreign package, I got the following warning message: Unrecognized record type 7, subtype 16 encountered in system file Thanks in advance for the help and apologies for the trouble, Celso [[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.htmlhttp://www.r-project.org/posting-guide.html [[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] parametric proportional hazard regression
On Fri, 7 Jul 2006, Valentin Dimitrov wrote: I am trying to find a suitable R-function for parametric proportional hazard regressions. The package survival contains the coxph() function which performs a Cox regression which leaves the base hazard unspecified, i.e. it is a semi-parametric method. The package Design contains the function pphsm() which is good for parametric proportional hazard regressions when the underlying base distribution is weibull or exponential. But what if I need a parametric proportional hazard model with the other usual distributions like 'gaussian', 'logistic', 'lognormal'and 'loglogistic? Those are not proportional hazards families of distributions. That is, if the distribution is gaussian for one value of the hazard ratio parameters it will not be gaussian for any other value. You can get accelerated failure models with these distributions using survreg() in the survival package. -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ 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] parametric proportional hazard regression
Valentin, Have you tried survreg() in the Design library? Regards, -Cody -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Valentin Dimitrov Sent: Friday, July 07, 2006 10:1 AM To: r-help@stat.math.ethz.ch Subject: [R] parametric proportional hazard regression Dear all, I am trying to find a suitable R-function for parametric proportional hazard regressions. The package survival contains the coxph() function which performs a Cox regression which leaves the base hazard unspecified, i.e. it is a semi-parametric method. The package Design contains the function pphsm() which is good for parametric proportional hazard regressions when the underlying base distribution is weibull or exponential. But what if I need a parametric proportional hazard model with the other usual distributions like 'gaussian', 'logistic', 'lognormal'and 'loglogistic? Thanks a lot for your support! Valentin Dimitrov Statistics and Econometrics University of Saarland __ 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 This e-mail, facsimile, or letter and any files or attachments transmitted with it contains information that is confidential and privileged. This information is intended only for the use of the individual(s) and entity(ies) to whom it is addressed. If you are the intended recipient, further disclosures are prohibited without proper authorization. If you are not the intended recipient, any disclosure, copying, printing, or use of this information is strictly prohibited and possibly a violation of federal or state law and regulations. If you have received this information in error, please notify Baylor Health Care System immediately at 1-866-402-1661 or via e-mail at [EMAIL PROTECTED] Baylor Health Care System, its subsidiaries, and affiliates hereby claim all applicable privileges related to this information. __ 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] Rpad Server Installation on Windows XP
Hello, I have a question about Rpad Server Installation on Windows XP. Here is a guide http://www.rpad.org/Rpad/ServerNotes.html and I have been reading this more than ten times to figure this out, but still having trouble executing a Rpad website properly. Problem: for example, if I go to http://loclhost/Rpad/example1.rpad, I can see the whole website with the page loading sign staying on my screen (like forever) but if I push calculate button noting happens and the page loading sign goes away. Here, I think I have completed all steps needed, but I think I got something wrong with Install the Statistice- R_perl_interface that come with Rpad in the serverversion directory. part. My Question: When I try installing serverversion files, I used -perl makefile.pl -nmake -nmake install in C:/Perl/bin directory. But I think the directory of perl might be some problem. I asked somebody out a week ago and he said I have to install perl on my path. Actually I don't quite know which directory he is talking about. I'm using xampp advanced version, so I should have perl installed and also I have active perl installed on my computer (c:/perl) What am I getting wrong? Question2: Do I have to install rterm too? If so, where? Any of your advice will be very appreciated. Thank you for your reading and have a great day everybody! best, Claire Ahn MIT __ 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 change the type of segments ends?
On Fri, 2006-07-07 at 11:47 -0400, Lu, Jiang Jane wrote: Hi, I am trying to plot odds ratios and the corresponding confidence intervals in horizontal segments. It would be ideal if the confidence interval segment can be drawn with little vertical bars at both ends. I have tried very hard to change the type of ends by using 'lend' arguments, but cannot make it. I even tried 'arrows()', but still failed. Following is the code I use: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x=drug.or,y=yaxis,type='p',pch=17,xlim=c(0,3),axes=FALSE, xlab='Odds Ratio',ylab='',main='Reference Group: A only') axis(1,at=seq(0,3,by=0.5),labels=paste(seq(0,3,by=0.5))) axis(2,at=yaxis,las=2) segments(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,col=4,lend=2) # or try #arrows(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,length=0.1,angle=0,cod e=3,col=4,lend=2) box() = Try this using arrows(): drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1, 4, by=1) plot(drug.or, yaxis ,type='p', pch=17, xlim=c(0,3), axes=FALSE, xlab='Odds Ratio', ylab='', main='Reference Group: A only') axis(1, at=seq(0,3,by=0.5), labels=paste(seq(0,3,by=0.5))) axis(2, at=yaxis, las=2) arrows(drug.orl, yaxis, drug.oru, yaxis, angle=90, length=0.1, code=3, col=4, lend=2) You need to specify an angle of 90 degrees to get the arrow heads to be perpendicular to the primary line segment. An angle of 0 superimposes the heads on the primary line segment, so they don't show. HTH, Marc Schwartz __ 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 change the type of segments ends?
You need angle=90. From the help page: angle: angle from the shaft of the arrow to the edge of the arrow head. On Fri, 7 Jul 2006, Lu, Jiang Jane wrote: Hi, I am trying to plot odds ratios and the corresponding confidence intervals in horizontal segments. It would be ideal if the confidence interval segment can be drawn with little vertical bars at both ends. I have tried very hard to change the type of ends by using 'lend' arguments, but cannot make it. I even tried 'arrows()', but still failed. Following is the code I use: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x=drug.or,y=yaxis,type='p',pch=17,xlim=c(0,3),axes=FALSE, xlab='Odds Ratio',ylab='',main='Reference Group: A only') axis(1,at=seq(0,3,by=0.5),labels=paste(seq(0,3,by=0.5))) axis(2,at=yaxis,las=2) segments(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,col=4,lend=2) # or try #arrows(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,length=0.1,angle=0,cod e=3,col=4,lend=2) box() = Any comments or suggestions would be greatly appreciated. Jane University of Pittsburgh Pittsburgh, PA 15261 __ 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 -- 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] How to change the type of segments ends?
Lu, Jiang Jane wrote: Hi, I am trying to plot odds ratios and the corresponding confidence intervals in horizontal segments. It would be ideal if the confidence interval segment can be drawn with little vertical bars at both ends. I have tried very hard to change the type of ends by using 'lend' arguments, but cannot make it. I even tried 'arrows()', but still failed. Following is the code I use: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x=drug.or,y=yaxis,type='p',pch=17,xlim=c(0,3),axes=FALSE, xlab='Odds Ratio',ylab='',main='Reference Group: A only') axis(1,at=seq(0,3,by=0.5),labels=paste(seq(0,3,by=0.5))) axis(2,at=yaxis,las=2) segments(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,col=4,lend=2) # or try #arrows(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,length=0.1,angle=0,cod e=3,col=4,lend=2) The call to arrows() seems to do what you want if angle=90. Also the lend argument seems unnecessary if you have angle=90. box() = Any comments or suggestions would be greatly appreciated. Jane University of Pittsburgh Pittsburgh, PA 15261 __ 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 -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 512-0171 (M, W, F) fax: (917) 438-0894 __ 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] dotplot (lattice) with panel.segments and groups
It seems to work if you handle the col= argument explicitly. Suggest you double check this since I have not carefully done so: barley$yield2 - with(barley, yield + 5) dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel=function(x, y, subscripts, groups, yield2, col, ...) { panel.dotplot(x, y, ...) panel.segments(x, as.numeric(y), yield2[subscripts], as.numeric(y), col = col[groups[subscripts]], ...) }) On 7/7/06, Sebastian Luque [EMAIL PROTECTED] wrote: Hi, The following produces almost exactly what I needed. The problems are that the 'panel.dotplot' call (commented) generates the error 'Error in NextMethod([) : argument subscripts is missing, with no default'. The other problem is that the colors alternate between the levels of the 'site' variable, rather than 'year'. barley$yield2 - with(barley, yield + 5) dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel=function(x, y, subscripts, yield2, ...) { ## panel.dotplot(x, y, ...) panel.segments(x, as.numeric(y), yield2[subscripts], as.numeric(y), ...) }) R sessionInfo() Version 2.3.1 (2006-06-01) i486-pc-linux-gnu attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: chron gmodels lattice 2.3-3 2.12.0 0.13-8 Can somebody please suggest how to properly make that 'panel.dotplot' call and set the 'col' argument? Thanks in advance. Cheers, -- Seb __ 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 change the type of segments ends?
On Fri, 2006-07-07 at 11:47 -0400, Lu, Jiang Jane wrote: Hi, I am trying to plot odds ratios and the corresponding confidence intervals in horizontal segments. It would be ideal if the confidence interval segment can be drawn with little vertical bars at both ends. I have tried very hard to change the type of ends by using 'lend' arguments, but cannot make it. I even tried 'arrows()', but still failed. Following is the code I use: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x=drug.or,y=yaxis,type='p',pch=17,xlim=c(0,3),axes=FALSE, xlab='Odds Ratio',ylab='',main='Reference Group: A only') axis(1,at=seq(0,3,by=0.5),labels=paste(seq(0,3,by=0.5))) axis(2,at=yaxis,las=2) segments(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,col=4,lend=2) # or try #arrows(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,length=0.1,angle=0,cod e=3,col=4,lend=2) box() = Any comments or suggestions would be greatly appreciated. Jane Hi Jane, Thanks for the reproducible example. Why do you think lend will do what you want? From ?par: 'lend' The line end style. This can be specified as an integer or string: '0' and 'round' mean rounded line caps [_default_]; '1' and 'butt' mean butt line caps; '2' and 'square' mean square line caps. As from R 2.3.0 this can be specified inline. This is talking about line endings - do you want hard or soft line ends or type of joins in line segments? It does not mean the type of line endings you might get in Adobe Illustrator or Inkscape vector drawing packages. You were close with arrows: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x = drug.or, y = yaxis, pch=17, xlim = c(0,3), axes=FALSE, xlab = 'Odds Ratio', ylab = '', main = 'Reference Group: A only') labs - seq(0, 3, by = 0.5) axis(1, at = labs, labels = labs) axis(2, at = yaxis, las = 2) arrows(x0 = drug.orl, x1 = drug.oru, y0 = yaxis, y1 = yaxis, length=0.1, code = 3, col = 4, angle = 90) the angle argument is the key. See arrows? You don't need lend at all to do what you want. HTH G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ 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 change the type of segments ends?
Thank you all for such a prompt response. I misunderstood the meaning of 'lend' in par(). Now I have got the ideal confidence segments. I appreciate all the specific explanations you have offered. Sincerely, Jane -Original Message- From: Gavin Simpson [mailto:[EMAIL PROTECTED] Sent: Friday, July 07, 2006 12:24 PM To: Lu, Jiang Jane Cc: r-help@stat.math.ethz.ch Subject: Re: [R] How to change the type of segments ends? On Fri, 2006-07-07 at 11:47 -0400, Lu, Jiang Jane wrote: Hi, I am trying to plot odds ratios and the corresponding confidence intervals in horizontal segments. It would be ideal if the confidence interval segment can be drawn with little vertical bars at both ends. I have tried very hard to change the type of ends by using 'lend' arguments, but cannot make it. I even tried 'arrows()', but still failed. Following is the code I use: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x=drug.or,y=yaxis,type='p',pch=17,xlim=c(0,3),axes=FALSE, xlab='Odds Ratio',ylab='',main='Reference Group: A only') axis(1,at=seq(0,3,by=0.5),labels=paste(seq(0,3,by=0.5))) axis(2,at=yaxis,las=2) segments(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,col=4,lend=2) # or try #arrows(x0=drug.orl,x1=drug.oru,y0=yaxis,y1=yaxis,length=0.1,angle=0,cod e=3,col=4,lend=2) box() = Any comments or suggestions would be greatly appreciated. Jane Hi Jane, Thanks for the reproducible example. Why do you think lend will do what you want? From ?par: 'lend' The line end style. This can be specified as an integer or string: '0' and 'round' mean rounded line caps [_default_]; '1' and 'butt' mean butt line caps; '2' and 'square' mean square line caps. As from R 2.3.0 this can be specified inline. This is talking about line endings - do you want hard or soft line ends or type of joins in line segments? It does not mean the type of line endings you might get in Adobe Illustrator or Inkscape vector drawing packages. You were close with arrows: drug.or - c(1.017,1.437,1.427,2.211) drug.orl - c(0.715,1.075,1.103,1.696) drug.oru - c(1.446,1.922,1.845,2.882) yaxis - seq(1,4,by=1) plot(x = drug.or, y = yaxis, pch=17, xlim = c(0,3), axes=FALSE, xlab = 'Odds Ratio', ylab = '', main = 'Reference Group: A only') labs - seq(0, 3, by = 0.5) axis(1, at = labs, labels = labs) axis(2, at = yaxis, las = 2) arrows(x0 = drug.orl, x1 = drug.oru, y0 = yaxis, y1 = yaxis, length=0.1, code = 3, col = 4, angle = 90) the angle argument is the key. See arrows? You don't need lend at all to do what you want. HTH G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ 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] replace values in data frame
The format is like this. Plot species Value P1 ACERRUB 3 P2 MAGNVIR2 P3 ARONARB 2 etc. imported using x-read.table(file=filename.txt) I want to replace a list of values in the 2nd column with another list. For example, I want to replace ARONARB with PHOTPYR. list-read.table(file=originalnames.txt) replace-read.table(file=replacementnames.txt) I tried to use replace in this manner: newx-replace(x,list,replace) however, I get the error message: error in replace, invalid subscript type. I have tried transforming the above list and modifying the column names (column 2), but to no avail. I hope this clarifies a little. Sorry about that. On 7/7/06, Marc Schwartz (via MN) [EMAIL PROTECTED] wrote: On Fri, 2006-07-07 at 11:20 -0400, Wade Wall wrote: Hi all, I have a three columned list that I have imported into R. The first column is a plot (ex. Plot1), the second is a species name (ex ACERRUB) and the third a numeric value. I want to replace some of the second column names with other names (for example replace ACERRUB with ACERDRU). The original and replacement values are in separate lists (not vectors), but I can't seem to find the right function to perform this. The replace function seems to only want to work with numbers. Any clues? Wade Without seeing the code you are using, we can only guess a syntax error of some sort. It works fine using the iris dataset: head(iris) Sepal.Length Sepal.Width Petal.Length Petal.Width Species 1 5.1 3.5 1.4 0.2 setosa 2 4.9 3.0 1.4 0.2 setosa 3 4.7 3.2 1.3 0.2 setosa 4 4.6 3.1 1.5 0.2 setosa 5 5.0 3.6 1.4 0.2 setosa 6 5.4 3.9 1.7 0.4 setosa 7 4.6 3.4 1.4 0.3 setosa 8 5.0 3.4 1.5 0.2 setosa 9 4.4 2.9 1.4 0.2 setosa 10 4.9 3.1 1.5 0.1 setosa iris$Species - replace(iris$Species, iris$Species == setosa, NewValue) head(iris) Sepal.Length Sepal.Width Petal.Length Petal.Width Species 1 5.1 3.5 1.4 0.2 NewValue 2 4.9 3.0 1.4 0.2 NewValue 3 4.7 3.2 1.3 0.2 NewValue 4 4.6 3.1 1.5 0.2 NewValue 5 5.0 3.6 1.4 0.2 NewValue 6 5.4 3.9 1.7 0.4 NewValue 7 4.6 3.4 1.4 0.3 NewValue 8 5.0 3.4 1.5 0.2 NewValue 9 4.4 2.9 1.4 0.2 NewValue 10 4.9 3.1 1.5 0.1 NewValue HTH, Marc Schwartz [[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] Converting data frame to zoo
Dear list, I know this is really basic question but I just couldn't get anything to work. (I did a R site search with keywords zoo and data frame but the server timed out on me.) I have a time series which has the following (typical) format, DATE Open High Low Close Volume 01-JAN-2006 5.25 5.25 5.25 5.25 256 I read the data in from a csv file with read.csv() and it defaulted to a data frame which I thought is fine. Now I want to convert it to zoo so I did x - zoo(my.df) which works just fine. But the Date column has been turned into a factor. Is there a way to make it into a Date. I've tried, x$Date - as.Date(x$Date) but R complains that Error in fromchar(x) : character string is not in a standard unambiguous format Thanks in advance. Horace W. Tso __ 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] replace values in data frame
Thanks for re-posting onlist. My offlist reply: The 'list' argument, as per ?replace, needs to be a vector of one or more indices into another vector, generally the source vector argument 'x', not the actual values that you want to replace. 'list' can either be explicit integers as the indices, or the result of a logical operation as I used in my example. In addition, the objects 'list' and 'replace' in your code are not vectors, but they are data frames. That is the default type of object created when using read.table(). You can use str(list) and str(replace) to check this. The str() function returns the structure of an object. It is likely that you really want list$V1 and replace$V1 as the vectors of interest here. V1 thru Vx are the default column names created when using read.table() if no header row exists in the incoming file. In addition, note that replace() is not vectorized. It will not handle multiple search and replace arguments in a single call. See my example using the iris dataset. You would need to do each one separately. You can create a loop of one type or another to cycle through multiple arguments if you wish, where each cycle through the loop is a single call to replace() with a new set of values as appropriate for the arguments. Finally, you should generally not use R object or function names for user created objects. Both 'list' and 'replace' are such objects. R will generally be able to differentiate, but there is no guarantee and it will help you avoid code debugging headaches. HTH, Marc Schwartz On Fri, 2006-07-07 at 12:58 -0400, Wade Wall wrote: The format is like this. Plot species Value P1 ACERRUB 3 P2 MAGNVIR2 P3 ARONARB 2 etc. imported using x-read.table(file=filename.txt) I want to replace a list of values in the 2nd column with another list. For example, I want to replace ARONARB with PHOTPYR. list-read.table(file=originalnames.txt) replace-read.table(file=replacementnames.txt) I tried to use replace in this manner: newx-replace(x,list,replace) however, I get the error message: error in replace, invalid subscript type. I have tried transforming the above list and modifying the column names (column 2), but to no avail. I hope this clarifies a little. Sorry about that. On Fri, 2006-07-07 at 11:20 -0400, Wade Wall wrote: Hi all, I have a three columned list that I have imported into R. The first column is a plot (ex. Plot1), the second is a species name (ex ACERRUB) and the third a numeric value. I want to replace some of the second column names with other names (for example replace ACERRUB with ACERDRU). The original and replacement values are in separate lists (not vectors), but I can't seem to find the right function to perform this. The replace function seems to only want to work with numbers. Any clues? Wade __ 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] Converting data frame to zoo
Check out read.zoo in the zoo package. On 7/7/06, Horace Tso [EMAIL PROTECTED] wrote: Dear list, I know this is really basic question but I just couldn't get anything to work. (I did a R site search with keywords zoo and data frame but the server timed out on me.) I have a time series which has the following (typical) format, DATE Open High Low Close Volume 01-JAN-2006 5.25 5.25 5.25 5.25 256 I read the data in from a csv file with read.csv() and it defaulted to a data frame which I thought is fine. Now I want to convert it to zoo so I did x - zoo(my.df) which works just fine. But the Date column has been turned into a factor. Is there a way to make it into a Date. I've tried, x$Date - as.Date(x$Date) but R complains that Error in fromchar(x) : character string is not in a standard unambiguous format Thanks in advance. Horace W. Tso __ 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-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] replace values in data frame
Wade, Given that you appear to have multiple search and replace items to deal with, here is a possible loop based Global Search and Replace solution: gsr - function(Source, Search, Replace) { if (length(Search) != length(Replace)) stop(Search and Replace Must Have Equal Number of Items\n) Changed - as.character(Source) for (i in 1:length(Search)) { cat(Replacing: , Search[i], With: , Replace[i], \n) Changed - replace(Changed, Changed == Search[i], Replace[i]) } cat(\n) Changed } Source: The source vector, which will be coerced to character Search: The Search values to be matched in Source as a character vector Replace: The values that will replace those found in 'Search' on a one-for-one basis. This function returns a character vector. As with replace(), the result must be assigned. The change is not done 'in place'. The function will also output the search and replace values to the console during the loop. Again, using the iris dataset as an example: iris$Species [1] setosa setosa setosa setosa setosa setosa [7] setosa setosa setosa setosa setosa setosa [13] setosa setosa setosa setosa setosa setosa [19] setosa setosa setosa setosa setosa setosa [25] setosa setosa setosa setosa setosa setosa [31] setosa setosa setosa setosa setosa setosa [37] setosa setosa setosa setosa setosa setosa [43] setosa setosa setosa setosa setosa setosa [49] setosa setosa versicolor versicolor versicolor versicolor [55] versicolor versicolor versicolor versicolor versicolor versicolor [61] versicolor versicolor versicolor versicolor versicolor versicolor [67] versicolor versicolor versicolor versicolor versicolor versicolor [73] versicolor versicolor versicolor versicolor versicolor versicolor [79] versicolor versicolor versicolor versicolor versicolor versicolor [85] versicolor versicolor versicolor versicolor versicolor versicolor [91] versicolor versicolor versicolor versicolor versicolor versicolor [97] versicolor versicolor versicolor versicolor virginica virginica [103] virginica virginica virginica virginica virginica virginica [109] virginica virginica virginica virginica virginica virginica [115] virginica virginica virginica virginica virginica virginica [121] virginica virginica virginica virginica virginica virginica [127] virginica virginica virginica virginica virginica virginica [133] virginica virginica virginica virginica virginica virginica [139] virginica virginica virginica virginica virginica virginica [145] virginica virginica virginica virginica virginica virginica Levels: setosa versicolor virginica Note that iris$Species is a factor with specific levels. The gsr() function above will coerce the Source argument to a character vector internally and return a character vector: iris$Species - gsr(iris$Species, c(setosa, versicolor, virginica), c(s1, v1, v2)) Replacing: setosa With: s1 Replacing: versicolor With: v1 Replacing: virginica With: v2 iris$Species [1] s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 [14] s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 [27] s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 [40] s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 s1 v1 v1 [53] v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 [66] v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 [79] v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 v1 [92] v1 v1 v1 v1 v1 v1 v1 v1 v1 v2 v2 v2 v2 [105] v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 [118] v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 [131] v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 v2 [144] v2 v2 v2 v2 v2 v2 v2 HTH, Marc Schwartz __ 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] Rpad Server Installation on Windows XP
Claire, I know you emailed me a question earlier and I just haven't gotten around to answering it. For both perl and rgui.exe (and rterm.exe) your path needs to state the location of these files. (I am not sure about this, but after you change the path you may even need to reboot for the change to be applied). Goto the dos prompt at C: and type in 'perl' and 'rterm' to make sure both are recognized. I don't really know what makefile does, but my understanding is that it just copies the files and you can do that yourself. Once you do that, your 'server' folder on the rapd server will have some .pl files. An example of the R_process.pl file is below. I don't know why there is a #! in front of the path on the first line, but make sure that path points to the location of your perl install. Your perl istall is likely different from the default in the .pl files. Once you have verified those two items, work through the troubleshooting portion of the rpad install notes and let us know which steps are working and which are not. As you can tell, I am no expert myself. I was very happy once I got it to work. Fortunately, it has worked like a charm ever since. I am just hoping I never have to do it again!! lol. #!c:/perl/bin/perl.exe # The following line is a test script to see if it works. # http://localhost/Rpad/server/R_process.pl?ID=ddNTlmHSvWZFcommand=R_commandsR_commands=print('hello' ) use Statistics::Rpad ; use strict ; use CGI qw/:standard send_http_header/; use Cwd ; use utf8; # the decoding stuff is from spell-check-logic.cgi in the htmlarea spellcheck plugin use Encode; On 7/7/06, Minjeong Ahn [EMAIL PROTECTED] wrote: Hello, I have a question about Rpad Server Installation on Windows XP. Here is a guide http://www.rpad.org/Rpad/ServerNotes.html and I have been reading this more than ten times to figure this out, but still having trouble executing a Rpad website properly. Problem: for example, if I go to http://loclhost/Rpad/example1.rpad, I can see the whole website with the page loading sign staying on my screen (like forever) but if I push calculate button noting happens and the page loading sign goes away. Here, I think I have completed all steps needed, but I think I got something wrong with Install the Statistice- R_perl_interface that come with Rpad in the serverversion directory. part. My Question: When I try installing serverversion files, I used -perl makefile.pl -nmake -nmake install in C:/Perl/bin directory. But I think the directory of perl might be some problem. I asked somebody out a week ago and he said I have to install perl on my path. Actually I don't quite know which directory he is talking about. I'm using xampp advanced version, so I should have perl installed and also I have active perl installed on my computer (c:/perl) What am I getting wrong? Question2: Do I have to install rterm too? If so, where? Any of your advice will be very appreciated. Thank you for your reading and have a great day everybody! best, Claire Ahn MIT __ 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 [[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] attach and detach question
I have a large R program that I am constantly running over and over again. At the beginning of this program, I create a hige matrix and a huge dataframe but these are constant. What I mean by constant is that, if I run the program over later, I really should just use the old matrix and dataframe ( if they exist ) that were created in a previous run so that the program doesn't have to spend time creating them. Unfortunately, I don't know how to do this so things take foreeer because every time i do a new run, i recreate these objects and the boss is getting a little annoyed. I know/think that I should use attach and detach commands but 1) i can't find an example somewhere of just saving two objects rather than the whole session. i've looked and looked and i can't find it. 2) if i am able to save these two objects, i was hoping that it would be possible to write code inside my R program that basically says, if these objects already exist in such and such database, then skip over this section of the code that creates them. if someone has some code or a page number or reference, this is fine because i used to do pretty much this in Splus so think i can figure it out if i can just find an example. thanks a lot. __ 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] Converting data frame to zoo
Thanks Gabor. I figured out what went wrong. The culprit turns out to be the headers in my data. read.zoo doesn't recognize column headers and complains Error in read.zoo(C:\\...\\table.csv, : index contains NAs Or is there an option as in read.table(x, header=...) ? After the header line is removed it works fine. H. Gabor Grothendieck [EMAIL PROTECTED] 7/7/2006 10:22 AM Check out read.zoo in the zoo package. On 7/7/06, Horace Tso [EMAIL PROTECTED] wrote: Dear list, I know this is really basic question but I just couldn't get anything to work. (I did a R site search with keywords zoo and data frame but the server timed out on me.) I have a time series which has the following (typical) format, DATE Open High Low Close Volume 01-JAN-2006 5.25 5.25 5.25 5.25 256 I read the data in from a csv file with read.csv() and it defaulted to a data frame which I thought is fine. Now I want to convert it to zoo so I did x - zoo(my.df) which works just fine. But the Date column has been turned into a factor. Is there a way to make it into a Date. I've tried, x$Date - as.Date(x$Date) but R complains that Error in fromchar(x) : character string is not in a standard unambiguous format Thanks in advance. Horace W. Tso __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Converting data frame to zoo
read.zoo supports all the arguments that read.table supports including header=. From the read.zoo help file: ...: further arguments passed to 'read.table'. Regards. On 7/7/06, Horace Tso [EMAIL PROTECTED] wrote: Thanks Gabor. I figured out what went wrong. The culprit turns out to be the headers in my data. read.zoo doesn't recognize column headers and complains Error in read.zoo(C:\\...\\table.csv, : index contains NAs Or is there an option as in read.table(x, header=...) ? After the header line is removed it works fine. H. Gabor Grothendieck [EMAIL PROTECTED] 7/7/2006 10:22 AM Check out read.zoo in the zoo package. On 7/7/06, Horace Tso [EMAIL PROTECTED] wrote: Dear list, I know this is really basic question but I just couldn't get anything to work. (I did a R site search with keywords zoo and data frame but the server timed out on me.) I have a time series which has the following (typical) format, DATE Open High Low Close Volume 01-JAN-2006 5.25 5.25 5.25 5.25 256 I read the data in from a csv file with read.csv() and it defaulted to a data frame which I thought is fine. Now I want to convert it to zoo so I did x - zoo(my.df) which works just fine. But the Date column has been turned into a factor. Is there a way to make it into a Date. I've tried, x$Date - as.Date(x$Date) but R complains that Error in fromchar(x) : character string is not in a standard unambiguous format Thanks in advance. Horace W. Tso __ 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-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] attach and detach question
There is an example of saving just two objects in the example section of ?save To check whether an object of a given name exists see ?exists On 7/7/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I have a large R program that I am constantly running over and over again. At the beginning of this program, I create a hige matrix and a huge dataframe but these are constant. What I mean by constant is that, if I run the program over later, I really should just use the old matrix and dataframe ( if they exist ) that were created in a previous run so that the program doesn't have to spend time creating them. Unfortunately, I don't know how to do this so things take foreeer because every time i do a new run, i recreate these objects and the boss is getting a little annoyed. I know/think that I should use attach and detach commands but 1) i can't find an example somewhere of just saving two objects rather than the whole session. i've looked and looked and i can't find it. 2) if i am able to save these two objects, i was hoping that it would be possible to write code inside my R program that basically says, if these objects already exist in such and such database, then skip over this section of the code that creates them. if someone has some code or a page number or reference, this is fine because i used to do pretty much this in Splus so think i can figure it out if i can just find an example. thanks a lot. __ 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] attach and detach question
[EMAIL PROTECTED] wrote: I have a large R program that I am constantly running over and over again. At the beginning of this program, I create a hige matrix and a huge dataframe but these are constant. What I mean by constant is that, if I run the program over later, I really should just use the old matrix and dataframe ( if they exist ) that were created in a previous run so that the program doesn't have to spend time creating them. Unfortunately, I don't know how to do this so things take foreeer because every time i do a new run, i recreate these objects and the boss is getting a little annoyed. I know/think that I should use attach and detach commands but 1) i can't find an example somewhere of just saving two objects rather than the whole session. i've looked and looked and i can't find it. 2) if i am able to save these two objects, i was hoping that it would be possible to write code inside my R program that basically says, if these objects already exist in such and such database, then skip over this section of the code that creates them. You probably want save() and load() rather than attach() and detach(). # Save only a couple of objects save(huge.mat, huge.df, file=myfile.Rdata) # Load objects only if they do not exist if(!exists(huge.mat) !exists (huge.df)) load(myfile.Rdata) else {cat(\n, Object(s) Already Exist, \n)} ?save ?load ?exists if someone has some code or a page number or reference, this is fine because i used to do pretty much this in Splus so think i can figure it out if i can just find an example. thanks a lot. __ 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 -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 512-0171 (M, W, F) fax: (917) 438-0894 __ 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] Levels and GLM
I am using the as.factor command to use with glm. When I use the command maj - as.factor(data.logistic$Majors) maj I receive the following output: [1] M M N M M M M N N M M M N M M M M M M M M M M M N M N N M M N M M N M M M M M [40] N M N M M N M M M N M N M N M N N N M N M M M M M M N M N M M M M M N N M M M [79] M M M N N M M N M N M M M M M M M M M M M M M M M N M M M M M N M M M M M N M [118] M M M N M N N M M M M M M M M N M N M M M M M N M M M M N M M M N N M M M N M [157] M M M M M M M M M M M M M N M M N N M M N M M M M M M M M M M M M M N M N M M [196] M N M M M M M M M M N M M M M M M M M N M M M M M M M M M M M M M M N M M N N [235] M M M M M N M M M M M M N N M M N M M M M M M M M M M M M M M M M N M M M M N [274] N M M M M M M N M M M M M M M M M M N N M N M M M M M M M M M M N M N N M M M [313] M M M M M M M N M M M M M N M M M M M M M M M M M M M M M N M M M M M M M N M [352] M N M N M M N M M M M N M M M M M M M M M M N M M N N Levels: M N When I enter: logistic.glm - glm(data.logistic$X100.Yard.Average ~ data.logistic$Overall + maj, family=binomial) logistic.glm I receive the following output: Call: glm(formula = data.logistic$X100.Yard.Average ~ data.logistic$Overall + maj, family = binomial) Coefficients: (Intercept) data.logistic$Overall majN 2.38819 -0.02718 -0.18385 Degrees of Freedom: 377 Total (i.e. Null); 375 Residual Null Deviance: 514.5 Residual Deviance: 410.7AIC: 416.7 My question: Why is there no output for majM? Any help would be greatly appreciated __ 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-pkgs] ggplot: a new system for drawing graphics in R
I think the ggplot package is extremely promising. Parts of the dokumentation are very good alread, e.g I recently managed to write my first own grob function. One thing I am missing in the dokumentation is a little more detail on how to modify the colours of a plot. Specifically, if I have: xx-x y g bar 1 1 1 1 1 5 2 3 2 3 1 3 2 4 2 2 df=read.table(textConnection(xx),header=T);df ggplot(df, aes=list(y=y, x=factor(x),bar=bar))-p ggbar(p, aes=list(fill=g,barcolour=g+1), avoid=dodge, sort=TRUE) ... I get a beatiful barplot, but would like to change the colors of the bars. I tried ?scfill and did not understand too much of the help file. I think some more examples would do it. Timm hadley wickham schrieb: ggplot provides a new system for drawing graphics in R, based on the Grammar of Graphics. It combines the advantages of both base and lattice graphics: conditioning and shared axes are handled automatically, and you can still build up a plot step by step from multiple data sources. It also implements a more sophisticated multidimensional conditioning system and a consistent interface to map data to visual attributes. ggplot (along with reshape) received the John Chambers Award for Statistical Computing. ggplot is available now from CRAN (install.packages(ggplot)) and more information is available at my website (http://had.co.nz/ggplot) including copies of talks, examples, and a guide showing how to convert your existing lattice code. To get started I recommend you look at: * the introductory vignette: vignette(introduction, ggplot) * help for the quick plotting command: ?qplot * help for the full plotting commands: ?ggplot I want to provide great documentation, so if there is anything you think I am missing, please let me know. Regards, Hadley ___ R-packages mailing list [EMAIL PROTECTED] https://stat.ethz.ch/mailman/listinfo/r-packages __ 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] attach and detach question
On Fri, 2006-07-07 at 12:57 -0500, [EMAIL PROTECTED] wrote: snipped / 1) i can't find an example somewhere of just saving two objects rather than the whole session. i've looked and looked and i can't find it. ?save 2) if i am able to save these two objects, i was hoping that it would be possible to write code inside my R program that basically says, if these objects already exist in such and such database, then skip over this section of the code that creates them. Is this the kind of thing you mean? I used the FileExists function from package RandomFields as this does the file checking - you could do something similar with ?file.list and match the name but why reinvent the wheel. I flipped your situation 2 around, if you have some really big objects then it would make sense to generate the objects, save them out to files. From a new session, you are likely not to have objects, so perhaps better to check the file exists, if so load it, then check the right objects are there. Alter to what suits you best. ## create two objects to be saved obj1 - matrix(rnorm(1000), ncol = 10) obj2 - matrix(rnorm(1000), ncol = 10) ## save them save(obj1, obj2, file = tmp.file.RData) ## clean up rm(list = ls()) ls() ## load the required package require(RandomFields) ## load the saved objects if(FileExists(tmp.file.RData)) { load(tmp.file.RData) if(exists(obj1) exists(obj2)) { ## more code here summary(obj1) summary(obj2) } else stop(Required objects don't exist!) } else stop(Big objects not found, check file exists!) ## clean up rm(list = ls()) ## if file exists but not correct objects if(FileExists(tmp.file.RData)) { load(tmp.file.RData) ## simulate different object names obj3 - obj1 obj4 - obj2 rm(obj1, obj2) if(exists(obj1) exists(obj2)) { ## more code here summary(obj1) summary(obj2) } else stop(Required objects don't exist!) } else stop(Big objects not found, check file exists!) ## clean up rm(list = ls()) ## if file doesn't exists unlink(tmp.file.RData) if(FileExists(tmp.file.RData)) { load(tmp.file.RData) ## remove the objects obj3 - obj1 obj4 - obj2 rm(obj1, obj2) if(exists(obj1) exists(obj2)) { ## more code here summary(obj1) summary(obj2) } else stop(Required objects don't exist!) } else stop(Big objects not found, check file exists!) HTH, G -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [t] +44 (0)20 7679 0522 ECRC ENSIS, UCL Geography, [f] +44 (0)20 7679 0565 Pearson Building, [e] gavin.simpsonATNOSPAMucl.ac.uk Gower Street, London [w] http://www.ucl.ac.uk/~ucfagls/cv/ London, UK. WC1E 6BT. [w] http://www.ucl.ac.uk/~ucfagls/ %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ 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] panel ordering in nlme and augPred plots
I'm not sufficiently familiar with 'trellis' / 'lattice' to provide an easy, complete answer to your question, but I can explain the current behavior and provide a hack to get you what you want. With luck, someone else will suggest an improvement. First, let's decompose your lovely, self-contained example in a more informative way: fm - lme(Orthodont) ap - augPred(fm, level = 0:1) plot(ap, skip = rep(c(F,T), c(16, 2))) Next, let's examine the structure of 'ap': str(ap) Classes augPred and `data.frame': 2862 obs. of 4 variables: $ age : num 8 10 12 14 8 10 12 14 8 10 ... $ .groups : Ord.factor w/ 27 levels M16M05M02..: 15 15 15 15 3 3 3 3 7 7 ... ..- attr(*, label)= chr Subject snip ... The key here is that 'ap$.groups' is of class Ord.factor with levels as follows: levels(ap$.groups) [1] M16 M05 M02 M11 M07 M08 M03 M12 M13 M14 M09 M15 [13] M06 M04 M01 M10 F10 F09 F06 F01 F05 F07 F02 F08 [25] F03 F04 F11 These names appear in the same order here that they appear in the trellis display, reading the display from left to right and bottom to top -- NOT top to bottom. This also tells you one way to get around this: Change the order of the levels. We can do this as follows: ap2 - ap ap2$.groups - with(ap, ordered(as.character(.groups), sort(levels(.groups plot(ap2, skip = rep(c(F,T), c(11, 1))) Hope this helps. Spencer Graves Nathaniel Derby wrote: Hi, I'm new at this, I'm very confused, and I think I'm missing something important here. In our pet example we have this: fm - lme(Orthodont) plot(Orthodont) plot(augPred(fm, level = 0:1)) which gives us a trellis plot with the females above the males, starting with F03, F04, F11, F06, etc. I thought the point of this was to create an ordering where the females are ordered (F01, F02, F03, etc -- followed by the males being ordered). However, the solution given ... fm - lme(Orthodont) plot(Orthodont) plot(augPred(fm1, level = 0:1), skip = rep(c(F,T), c(16, 2))) ... doesn't solve it -- although it does do all the females before starting on the males. That is, it starts with F02, F08, F03, ... which isn't in order either. Running Petr's code also gave output which wasn't ordered by the subjects. Could someone please explain to me how to order the panels of the trellis plot by the subjects? thanks, Nandor __ 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] dotplot (lattice) with panel.segments and groups
On 7/7/06, Sebastian Luque [EMAIL PROTECTED] wrote: Hi, The following produces almost exactly what I needed. The problems are that the 'panel.dotplot' call (commented) generates the error 'Error in NextMethod([) : argument subscripts is missing, with no default'. It's just as it says: panel.dotplot wants a 'subscripts' argument when 'groups' is not null, and you have forgotten to give it one. The other problem is that the colors alternate between the levels of the 'site' variable, rather than 'year'. It's doing what it's being told to do. You probably want something like dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel = panel.superpose, panel.groups = function(x, y, subscripts, yield2, col, ...) { panel.xyplot(x, y, col = col, ...) panel.segments(x, y, yield2[subscripts], y, col = col) }) Note that you don't want to use panel.dotplot here as it would draw a reference line both times. If you do want a reference line, add that in the panel function. The explicit use of 'col' should not have been necessary, i.e., the following should have worked: dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel = panel.superpose, panel.groups = function(x, y, subscripts, yield2, ...) { panel.xyplot(x, y, col = ...) panel.segments(x, y, yield2[subscripts], y, ...) }) I believe this doesn't work because of a grid bug. I'll take that up in a separate thread. -Deepayan barley$yield2 - with(barley, yield + 5) dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel=function(x, y, subscripts, yield2, ...) { ## panel.dotplot(x, y, ...) panel.segments(x, as.numeric(y), yield2[subscripts], as.numeric(y), ...) }) R sessionInfo() Version 2.3.1 (2006-06-01) i486-pc-linux-gnu attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: chron gmodels lattice 2.3-3 2.12.0 0.13-8 Can somebody please suggest how to properly make that 'panel.dotplot' call and set the 'col' argument? Thanks in advance. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] dotplot (lattice) with panel.segments and groups
Could you explain what panel.groups= does and what the difference is between panel.groups= and panel= ? In ?xyplot it just says: panel.groups: useful mostly for 'xyplot' and 'densityplot'. Applies when 'panel' is 'panel.superpose' (which happens by default in these cases if 'groups' is non-null) which indicates when it might apply but not what it does. Also can we assume from the above that the panel=panel.superpose in your response could have been omitted? Thanks. On 7/7/06, Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/7/06, Sebastian Luque [EMAIL PROTECTED] wrote: Hi, The following produces almost exactly what I needed. The problems are that the 'panel.dotplot' call (commented) generates the error 'Error in NextMethod([) : argument subscripts is missing, with no default'. It's just as it says: panel.dotplot wants a 'subscripts' argument when 'groups' is not null, and you have forgotten to give it one. The other problem is that the colors alternate between the levels of the 'site' variable, rather than 'year'. It's doing what it's being told to do. You probably want something like dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel = panel.superpose, panel.groups = function(x, y, subscripts, yield2, col, ...) { panel.xyplot(x, y, col = col, ...) panel.segments(x, y, yield2[subscripts], y, col = col) }) Note that you don't want to use panel.dotplot here as it would draw a reference line both times. If you do want a reference line, add that in the panel function. The explicit use of 'col' should not have been necessary, i.e., the following should have worked: dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel = panel.superpose, panel.groups = function(x, y, subscripts, yield2, ...) { panel.xyplot(x, y, col = ...) panel.segments(x, y, yield2[subscripts], y, ...) }) I believe this doesn't work because of a grid bug. I'll take that up in a separate thread. -Deepayan barley$yield2 - with(barley, yield + 5) dotplot(site ~ yield | variety, data=barley, groups=year, yield2=barley$yield2, col=c(gray, black), panel=function(x, y, subscripts, yield2, ...) { ## panel.dotplot(x, y, ...) panel.segments(x, as.numeric(y), yield2[subscripts], as.numeric(y), ...) }) R sessionInfo() Version 2.3.1 (2006-06-01) i486-pc-linux-gnu attached base packages: [1] methods stats graphics grDevices utils datasets [7] base other attached packages: chron gmodels lattice 2.3-3 2.12.0 0.13-8 Can somebody please suggest how to properly make that 'panel.dotplot' call and set the 'col' argument? Thanks in advance. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Multistage Sampling
Dear WizaRds, dear Thomas, First of all, I want to tell you how grateful I am for all your support. I wish I will be able to help others along one day the same way you do. Thank you so much. I am struggling with a multistage sampling design: library(survey) multi3 - data.frame(cluster=c(1,1,1,1 ,2,2,2, 3,3), id=c(1,2,3,4, 1,2,3, 1,2), nl=c(4,4,4,4, 3,3,3, 2,2), Nl=c(100,100,100,100, 50,50,50, 75,75), M=rep(23,9), y=c(23,33,77,25, 35,74,27, 37,72) ) dmulti3 - svydesign(id=~cluster+id, fpc=~M+Nl, data=multi3) svymean (~y, dmulti3) mean SE y 45.796 5.5483 svytotal(~y, dmulti3) totalSE y 78999 13643 and I estimate the population total as N=M/m sum(Nl) = 23/3*(100+50+75)=1725. With this, my variance estimator is: y1-mean(multi3$y[1:4]) # 39.5 y2-mean(multi3$y[5:7]) # 45.33 y3-mean(multi3$y[8:9]) # 54.5 yT1-100*y1 # 3950 total cluster 1 yT2-50*y2 # 2266.67 total cluster 2 yT3-75*y3 # 4087.5 total cluster 3 ybarT-1/3*sum(yT1,yT2,yT3) # 3434.722 s1 - var(multi3$y[1:4]) # 643.67 var cluster 1 s2 - var(multi3$y[5:7]) # 632.33 var cluster 2 s3 - var(multi3$y[8:9]) # 612.5 var cluster 3 var.yT - 23^2*( 20/23*1/6*sum( (yT1-ybarT)^2,(yT2-ybarT)^2,(yT3-ybarT)^2 ) + 1/69 * sum(100*96*s1, 50*47*s2, 75*73*s3) ) # 242 101 517 but var.yT/1725^2 = 81.36157 SE = 9.02006, but it should be SE=13643/1725=7.90899 Is this calculation correct? I remember svytotal using a different variance estimator compared to svymean, and that svytotal gives the unbiased estimation. To solve the problem, I went ahead and tried to calibrate the design object, telling Survey the population total N=1725: dmulti3.cal - calibrate(dmulti3, ~1, pop=1725) svymean (~y, dmulti3.cal) mean SE y 45.796 5.5483 svytotal(~y, dmulti3.cal) total SE y 78999 9570.7 , which indeed gives me the computed svymean SE, but alas, I still don't know why my variance is so different. I think it might have sthg to do with a differently computed N and the fact that your estimator formula is a different one. Since I calculated the Taylor Series solution, i suppose there must be another approach? The calibration help page tells me to enter a list of population total vectors for each cluster, which would result in: dmulti3.cal - calibrate(dmulti3, ~1, pop=c(100,50,75)) Error in regcalibrate.survey.design2(design, formula, population, aggregate.stage = aggregate.stage, : Population and sample totals are not the same length. I am very grateful for your help and wish you alle the best Yours mark __ 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] Multistage Sampling
From: Mark Hempelmann [EMAIL PROTECTED] Date: Fri Jul 07 14:05:29 CDT 2006 To: r-help@stat.math.ethz.ch Subject: [R] Multistage Sampling i also find it an truly amazing group also. the general kindness and generosity of everyone is beyond belief. it will be a long time coming but i hope i can help some day also. unfortunately, i can't help you with your question either. mark Dear WizaRds, dear Thomas, First of all, I want to tell you how grateful I am for all your support. I wish I will be able to help others along one day the same way you do. Thank you so much. I am struggling with a multistage sampling design: library(survey) multi3 - data.frame(cluster=c(1,1,1,1 ,2,2,2, 3,3), id=c(1,2,3,4, 1,2,3, 1,2), nl=c(4,4,4,4, 3,3,3, 2,2), Nl=c(100,100,100,100, 50,50,50, 75,75), M=rep(23,9), y=c(23,33,77,25, 35,74,27, 37,72) ) dmulti3 - svydesign(id=~cluster+id, fpc=~M+Nl, data=multi3) svymean (~y, dmulti3) mean SE y 45.796 5.5483 svytotal(~y, dmulti3) totalSE y 78999 13643 and I estimate the population total as N=M/m sum(Nl) = 23/3*(100+50+75)=1725. With this, my variance estimator is: y1-mean(multi3$y[1:4]) # 39.5 y2-mean(multi3$y[5:7]) # 45.33 y3-mean(multi3$y[8:9]) # 54.5 yT1-100*y1 # 3950 total cluster 1 yT2-50*y2 # 2266.67 total cluster 2 yT3-75*y3 # 4087.5 total cluster 3 ybarT-1/3*sum(yT1,yT2,yT3) # 3434.722 s1 - var(multi3$y[1:4]) # 643.67 var cluster 1 s2 - var(multi3$y[5:7]) # 632.33 var cluster 2 s3 - var(multi3$y[8:9]) # 612.5 var cluster 3 var.yT - 23^2*( 20/23*1/6*sum( (yT1-ybarT)^2,(yT2-ybarT)^2,(yT3-ybarT)^2 ) + 1/69 * sum(100*96*s1, 50*47*s2, 75*73*s3) ) # 242 101 517 but var.yT/1725^2 = 81.36157 SE = 9.02006, but it should be SE=13643/1725=7.90899 Is this calculation correct? I remember svytotal using a different variance estimator compared to svymean, and that svytotal gives the unbiased estimation. To solve the problem, I went ahead and tried to calibrate the design object, telling Survey the population total N=1725: dmulti3.cal - calibrate(dmulti3, ~1, pop=1725) svymean (~y, dmulti3.cal) mean SE y 45.796 5.5483 svytotal(~y, dmulti3.cal) total SE y 78999 9570.7 , which indeed gives me the computed svymean SE, but alas, I still don't know why my variance is so different. I think it might have sthg to do with a differently computed N and the fact that your estimator formula is a different one. Since I calculated the Taylor Series solution, i suppose there must be another approach? The calibration help page tells me to enter a list of population total vectors for each cluster, which would result in: dmulti3.cal - calibrate(dmulti3, ~1, pop=c(100,50,75)) Error in regcalibrate.survey.design2(design, formula, population, aggregate.stage = aggregate.stage, : Population and sample totals are not the same length. I am very grateful for your help and wish you alle the best Yours mark __ 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] dotplot (lattice) with panel.segments and groups
On 7/7/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: Could you explain what panel.groups= does and what the difference is between panel.groups= and panel= ? In ?xyplot it just says: panel.groups: useful mostly for 'xyplot' and 'densityplot'. Applies when 'panel' is 'panel.superpose' (which happens by default in these cases if 'groups' is non-null) which indicates when it might apply but not what it does. That's wrong (it used to be right - a good example of why \synopsis is bad). Since lattice 0.13-x, panel.superpose is never the default panel function. An updated version with improved documentation should be out soon. 'panel.groups' is simply an argument to panel.superpose, and is described in ?panel.superpose. Thus, it only makes sense as an argument to xyplot/dotplot/whatever when the panel function is panel.superpose, and not otherwise. The entry for the graphical parameters in ?panel.superpose isn't as useful as it could be, I have just updated it to read: col, col.line, col.symbol, pch, cex, fill, font, fontface, fontfamily, lty, lwd, alpha: graphical parameters, replicated to be as long as the number of groups. These are eventually passed down to 'panel.groups', but as scalars rather than vectors. When 'panel.groups' is called for the i-th level of 'groups', the corresponding element of each graphical parameter is passed to it. Hope that makes things a bit clearer. Deepayan __ 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] FW: Levels and GLM
One correction... since you are fitting a logistic model, it is technically correct to say the mean value of the linear predictor, instead of mean response. 20 lashes for me. Max -Original Message- From: Kuhn, Max Sent: Friday, July 07, 2006 4:11 PM To: 'r-help@stat.math.ethz.ch' Subject: [R] Levels and GLM jdrapp, By default, R fits full rank models. If you are coming from SAS, you're probably used to less than full rank model parameterizations. From Section 11.1.1 of An Introduction to R at http://cran.r-project.org/doc/manuals/R-intro.html#Contrasts there is this: What about a k-level factor A? The answer differs for unordered and ordered factors. For unordered factors k - 1 columns are generated for the indicators of the second, ..., kth levels of the factor. (Thus the implicit parameterization is to contrast the response at each level with that at the first.) So level M is the reference cell. Assuming that data.logistic$Overall is continuous, the intercept is the estimate of the mean response when maj = M and data.logistic$Overall = 0. The estimate for majN is the difference between the reference cell (estimated by the intercept) and the mean response when maj = N and data.logistic$Overall = 0. You should check out ?model.matrix and ?contrasts. Max I am using the as.factor command to use with glm. When I use the command maj - as.factor(data.logistic$Majors) maj I receive the following output: [1] M M N M M M M N N M M M N M M M M M M M M M M M N M N N M M N M M N M M M M M [40] N M N M M N M M M N M N M N M N N N M N M M M M M M N M N M M M M M N N M M M [79] M M M N N M M N M N M M M M M M M M M M M M M M M N M M M M M N M M M M M N M [118] M M M N M N N M M M M M M M M N M N M M M M M N M M M M N M M M N N M M M N M [157] M M M M M M M M M M M M M N M M N N M M N M M M M M M M M M M M M M N M N M M [196] M N M M M M M M M M N M M M M M M M M N M M M M M M M M M M M M M M N M M N N [235] M M M M M N M M M M M M N N M M N M M M M M M M M M M M M M M M M N M M M M N [274] N M M M M M M N M M M M M M M M M M N N M N M M M M M M M M M M N M N N M M M [313] M M M M M M M N M M M M M N M M M M M M M M M M M M M M M N M M M M M M M N M [352] M N M N M M N M M M M N M M M M M M M M M M N M M N N Levels: M N When I enter: logistic.glm - glm(data.logistic$X100.Yard.Average ~ data.logistic$Overall + maj, family=binomial) logistic.glm I receive the following output: Call: glm(formula = data.logistic$X100.Yard.Average ~ data.logistic$Overall + maj, family = binomial) Coefficients: (Intercept) data.logistic$Overall majN 2.38819 -0.02718 -0.18385 Degrees of Freedom: 377 Total (i.e. Null); 375 Residual Null Deviance:514.5 Residual Deviance: 410.7 AIC: 416.7 My question: Why is there no output for majM? Any help would be greatly appreciated -- LEGAL NOTICE\ Unless expressly stated otherwise, this messag...{{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] FW: Levels and GLM
Kuhn, Max [EMAIL PROTECTED] writes: One correction... since you are fitting a logistic model, it is technically correct to say the mean value of the linear predictor, instead of mean response. 20 lashes for me. ...plus five more for forgetting that the link is nonlinear and that you can't meaningfully speak about the mean on the linear predictor scale. Value of linear predictor corresponding to the mean response is correct (I hope...) So level M is the reference cell. Assuming that data.logistic$Overall is continuous, the intercept is the estimate of the mean response when maj = M and data.logistic$Overall = 0. The estimate for majN is the difference between the reference cell -- 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] dotplot (lattice) with panel.segments and groups
On 7/7/06, Deepayan Sarkar [EMAIL PROTECTED] wrote: On 7/7/06, Gabor Grothendieck [EMAIL PROTECTED] wrote: Could you explain what panel.groups= does and what the difference is between panel.groups= and panel= ? In ?xyplot it just says: panel.groups: useful mostly for 'xyplot' and 'densityplot'. Applies when 'panel' is 'panel.superpose' (which happens by default in these cases if 'groups' is non-null) which indicates when it might apply but not what it does. That's wrong (it used to be right - a good example of why \synopsis is bad). Since lattice 0.13-x, panel.superpose is never the default panel function. An updated version with improved documentation should be out soon. 'panel.groups' is simply an argument to panel.superpose, and is described in ?panel.superpose. Thus, it only makes sense as an argument to xyplot/dotplot/whatever when the panel function is panel.superpose, and not otherwise. The entry for the graphical parameters in ?panel.superpose isn't as useful as it could be, I have just updated it to read: col, col.line, col.symbol, pch, cex, fill, font, fontface, fontfamily, lty, lwd, alpha: graphical parameters, replicated to be as long as the number of groups. These are eventually passed down to 'panel.groups', but as scalars rather than vectors. When 'panel.groups' is called for the i-th level of 'groups', the corresponding element of each graphical parameter is passed to it. Hope that makes things a bit clearer. Deepayan Thanks. That does help. __ 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] Multistage Sampling
On Fri, 7 Jul 2006, Mark Hempelmann wrote: library(survey) multi3 - data.frame(cluster=c(1,1,1,1 ,2,2,2, 3,3), id=c(1,2,3,4, 1,2,3, 1,2), nl=c(4,4,4,4, 3,3,3, 2,2), Nl=c(100,100,100,100, 50,50,50, 75,75), M=rep(23,9), y=c(23,33,77,25, 35,74,27, 37,72) ) dmulti3 - svydesign(id=~cluster+id, fpc=~M+Nl, data=multi3) svymean (~y, dmulti3) mean SE y 45.796 5.5483 svytotal(~y, dmulti3) totalSE y 78999 13643 and I estimate the population total as N=M/m sum(Nl) = 23/3*(100+50+75)=1725. With this, my variance estimator is: y1-mean(multi3$y[1:4]) # 39.5 y2-mean(multi3$y[5:7]) # 45.33 y3-mean(multi3$y[8:9]) # 54.5 yT1-100*y1 # 3950 total cluster 1 yT2-50*y2 # 2266.67 total cluster 2 yT3-75*y3 # 4087.5 total cluster 3 ybarT-1/3*sum(yT1,yT2,yT3) # 3434.722 s1 - var(multi3$y[1:4]) # 643.67 var cluster 1 s2 - var(multi3$y[5:7]) # 632.33 var cluster 2 s3 - var(multi3$y[8:9]) # 612.5 var cluster 3 var.yT - 23^2*( 20/23*1/6*sum( (yT1-ybarT)^2,(yT2-ybarT)^2,(yT3-ybarT)^2 ) + 1/69 * sum(100*96*s1, 50*47*s2, 75*73*s3) ) # 242 101 517 I don't have any of my reference books here today, but if you use var.yT - 23^2*( 20/23*1/6*sum( (yT1-ybarT)^2,(yT2-ybarT)^2,(yT3-ybarT)^2 ) + 1/69 * sum(100*96*s1/4, 50*47*s2/3, 75*73*s3/2) ) # 242 101 517 the results agrees with svytotal(), and with Stata, and with formulas in a couple of sets of lecture notes I found by Googling. but var.yT/1725^2 = 81.36157 SE = 9.02006, but it should be SE=13643/1725=7.90899 Is this calculation correct? I remember svytotal using a different variance estimator compared to svymean, and that svytotal gives the unbiased estimation. This calculation is not correct for the mean, since it ignores the uncertainty in the estimated population total. The correct standard error comes from treating the mean as a ratio of estimated total to estimated population size. In this case you have to do it that way since you don't know the population size, but R always does it this way. Because the estimated population size and total are correlated, taking into account the uncertainty in the denominator actually reduces the standard error. The easiest way to reproduce the result that R gets is to do it the same way that R does: compute the standard error of the mean as the standard error of the total of a suitable set of estimating functions. If you define a new variable (y-45.796*1)/1725 and estimate the standard error of the total of this variable it will give: svytotal(~I((y-45.796)/1725),dmulti3) I((y - 45.796)/1725) 0.0002963 5.5482 which is what svymean() gives for the standard error of the mean of y. Using your formula for the variance of the total (with the corrections above) on this variable also gives sqrt(var.yT) [1] 5.54824 -thomas Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ 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] Harmonic Regression in R
Since I haven't seen an answer to this, I'll offer a couple of comments. I don't recall having heard the term 'harmonic regression' prior to seeing your email, but it sounded interesting, so I did some searching. First, RSiteSearch(harminic regression) produced 7 hits, one of which discusses the 'cyclones' package, which may be what you want: http://finzi.psych.upenn.edu/R/library/cyclones/html/00Index.html Second, for the benefit of folks like me who aren't sure of the definition, it appears to be linear regression on sines and cosines of something like 'time'. The following link describes how to do this using 'lm' in S-Plus or R: http://www.math.jmu.edu/~tomitayx/math328/Ch6SlideD.pdf Hope this helps. Spencer Graves Airon Yiu wrote: Dear all: Does anyone has harmonic regresssion analysis package written in R (to be used in Windows platform) ? Thanks ___ YM - Â÷½u°T®§ ´Nºâ§A¨S¦³¤Wºô¡A§AªºªB¤Í¤´¥i¥H¯d¤U°T®§µ¹§A¡A·í§A¤Wºô®É´N¯à¥ß§Y¬Ý¨ì¡A¥ô¦ó»¡¸Ü³£ÉN¨«¥¢¡C http://messenger.yahoo.com.hk [[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-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] Harmonic Regression in R
I have not seen it myself (I have the first edition which uses FORTRAN) but I believe the second edition of Peter Bloomfield's book on Fourier Analysis contains harmonic regression code in S-Plus and that may work in R. On 7/7/06, Spencer Graves [EMAIL PROTECTED] wrote: Since I haven't seen an answer to this, I'll offer a couple of comments. I don't recall having heard the term 'harmonic regression' prior to seeing your email, but it sounded interesting, so I did some searching. First, RSiteSearch(harminic regression) produced 7 hits, one of which discusses the 'cyclones' package, which may be what you want: http://finzi.psych.upenn.edu/R/library/cyclones/html/00Index.html Second, for the benefit of folks like me who aren't sure of the definition, it appears to be linear regression on sines and cosines of something like 'time'. The following link describes how to do this using 'lm' in S-Plus or R: http://www.math.jmu.edu/~tomitayx/math328/Ch6SlideD.pdf Hope this helps. Spencer Graves Airon Yiu wrote: Dear all: Does anyone has harmonic regresssion analysis package written in R (to be used in Windows platform) ? Thanks ___ YM - 離線訊息 就算你沒有上網,你的朋友仍可以留下訊息給你,當你上網時就能立即看到,任何說話都冇走失。 http://messenger.yahoo.com.hk [[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-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] KhmaladzeTest
Hello. I am a beginer in R and I can not implement the KhmaladzeTest in the following command. Please help me!!! PD: I attach thw results and the messages of the R program R : Copyright 2006, The R Foundation for Statistical Computing Version 2.3.1 (2006-06-01) ISBN 3-900051-07-0 R es un software libre y viene sin GARANTIA ALGUNA. Usted puede redistribuirlo bajo ciertas circunstancias. Escriba 'license()' o 'licence()' para detalles de distribucion. R es un proyecto colaborativo con muchos contribuyentes. Escriba 'contributors()' para obtener mas informacion y 'citation()' para saber como citar R o paquetes de R en publicaciones. Escriba 'demo()' para demostraciones, 'help()' para el sistema on-line de ayuda, o 'help.start()' para abrir el sistema de ayuda HTML con su navegador. Escriba 'q()' para salir de R. utils:::menuInstallLocal() package 'quantreg' successfully unpacked and MD5 sums checked updating HTML package descriptions utils:::menuInstallLocal() package 'foreign' successfully unpacked and MD5 sums checked updating HTML package descriptions utils:::menuInstallLocal() package 'Rcmdr' successfully unpacked and MD5 sums checked updating HTML package descriptions local({pkg - select.list(sort(.packages(all.available = TRUE))) + if(nchar(pkg)) library(pkg, character.only=TRUE)}) local({pkg - select.list(sort(.packages(all.available = TRUE))) + if(nchar(pkg)) library(pkg, character.only=TRUE)}) quantreg package loaded: To cite see citation(quantreg) local({pkg - select.list(sort(.packages(all.available = TRUE))) + if(nchar(pkg)) library(pkg, character.only=TRUE)}) local({pkg - select.list(sort(.packages(all.available = TRUE))) + if(nchar(pkg)) library(pkg, character.only=TRUE)}) Loading required package: tcltk Loading Tcl/Tk interface ... done --- Please select a CRAN mirror for use in this session --- also installing the dependencies 'acepack', 'scatterplot3d', 'fBasics', 'Hmisc', 'quadprog', 'oz', 'mlbench', 'randomForest', 'SparseM', 'xtable', 'chron', 'fCalendar', 'its', 'tseries', 'DAAG', 'e1071', 'mvtnorm', 'zoo', 'strucchange', 'sandwich', 'dynlm', 'leaps' probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/acepack_1.3-2.2.zip' Content type 'application/zip' length 55667 bytes URL abierta downloaded 54Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/scatterplot3d_0.3-24.zip' Content type 'application/zip' length 540318 bytes URL abierta downloaded 527Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/fBasics_221.10065.zip' Content type 'application/zip' length 3327499 bytes URL abierta downloaded 3249Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/Hmisc_3.0-12.zip' Content type 'application/zip' length 1993038 bytes URL abierta downloaded 1946Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/quadprog_1.4-8.zip' Content type 'application/zip' length 38626 bytes URL abierta downloaded 37Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/oz_1.0-13.zip' Content type 'application/zip' length 39640 bytes URL abierta downloaded 38Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/mlbench_1.1-1.zip' Content type 'application/zip' length 1324913 bytes URL abierta downloaded 1293Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/randomForest_4.5-16.zip' Content type 'application/zip' length 209710 bytes URL abierta downloaded 204Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/SparseM_0.68.zip' Content type 'application/zip' length 728594 bytes URL abierta downloaded 711Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/xtable_1.3-2.zip' Content type 'application/zip' length 56703 bytes URL abierta downloaded 55Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/chron_2.3-4.zip' Content type 'application/zip' length 101287 bytes URL abierta downloaded 98Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/fCalendar_221.10065.zip' Content type 'application/zip' length 754551 bytes URL abierta downloaded 736Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/its_1.1.1.zip' Content type 'application/zip' length 194287 bytes URL abierta downloaded 189Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/tseries_0.10-3.zip' Content type 'application/zip' length 392748 bytes URL abierta downloaded 383Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/DAAG_0.79.zip' Content type 'application/zip' length 796328 bytes URL abierta downloaded 777Kb probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/e1071_1.5-13.zip' Content type 'application/zip' length 627972 bytes URL abierta probando la URL 'http://cran.au.r-project.org/bin/windows/contrib/2.3/mvtnorm_0.7-2.zip' Content type
Re: [R] parametric proportional hazard regression
Those are not proportional hazards families of distributions. That is, if the distribution is gaussian for one value of the hazard ratio parameters it will not be gaussian for any other value. You can get accelerated failure models with these distributions using survreg() in the survival package. -thomas I do not need a accelerated failure model, but a proportional hazard model with a f0= weibull, exponential, loglogistic or lognormal baseline distribution. The hazard function is lambda(t)=exp(Xi*beta)*lambda0(t), where lambda0 is the baseline hazard lambda0(t)=f0(t)/(1-F0(t)) where f0 and F0 are the baseline density and cumulative distribution functions. This is a proportional hazard model since the ratio lambda(t|Xi)/lambda(t|Xj)=exp(Xi*beta)/exp(Xj*beta) does not depend on t. Valentin __ 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] parametric proportional hazard regression
Cody, I have tried the survreg() in the Design library, it is analogous to survreg() in the survival library and it seems to me it is designed only for accelerated time models like the accelerated failure model (or accelerated lifetime model) and not for proportional hazard models. Correct me if I am wrong. Valentin --- Hamilton, Cody [EMAIL PROTECTED] wrote: Valentin, Have you tried survreg() in the Design library? Regards, -Cody __ 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] Problems when computing the 1rst derivative of mixtures of densities
What problem are you trying to solve? The mixture proportion you are trying to estimate violates the assumptions that provide a normal approximation to the distribution of maximum likelihood estimators. These situations even violate the assumptions for 2*log(likelihood ratio) being approximately chi-square. If you want hypothesis tests or confidence intervals, Pinheiro and Bates (2000) recommended using 2*log(likelihood ratio) modified as suggested by Monte Carlo results. More recently, Bates has incorporated an 'mcmcsamp' function in the 'lme4' and 'Matrix' libraries. The best references I know on testing a parameter at a boundary are the following: Pinheiro and Bates (2000) Mixed-Effects Models in S and S-PLUS (Springer, sec. 2.4) Crainiceanu, Ruppert and Vogelsang (2003) “Some properties of Likelihood Ratio tests in linear mixed models” Crainiceanu, Ruppert, Claeskens, and Wand (2005) “Exact Likelihood Ratio Tests for Penalized Splines”, Biometrika, 92(1) Hope this helps. Spencer Graves p.s. See also inline. Clément Viel wrote: Hi everybody, I am currently working on mixtures of two densities ( f(xi,teta)= (1-teta)*f1(xi) + teta*f2(xi) ), particularly on the behavior of the variance for teta=0 (so sample only comes from the first distribution). To determine the maximum likelihood estimator I use the Newton-Rapdon Iteration. But when computing the first derivative I get a none linear function (with several asymptotes) which is completely absurd. Why do you think this is absurd? Nonlinear estimation can be very difficult. Also, have you reviewed the capabilities in R for working with mixtures of distributions? I just got 167 hits for RSiteSearch(mixtures) and 49 for RSiteSearch(mixture distributions, functions). This is my function to compute the first derivative: phy=function(teta,vect1,vect2){ return( sum(( vect2 - vect1) / (( 1 - teta) * vect1 + teta * vect2))) } note: vect1 and vect2 contains values of the two distributions computed from sample previously extracted. Beside, vect2 - vect1 is constant and ( 1 - teta) * vect1 + teta * vect2) is linear and always defined so I am expecting a linear function for the first derivative. To my mind, it's likely comes from the operation of division but I don't understand why results are skewed. Have you got any suggestions, please? __ 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] parametric proportional hazard regression
Valentin Dimitrov wrote: Those are not proportional hazards families of distributions. That is, if the distribution is gaussian for one value of the hazard ratio parameters it will not be gaussian for any other value. You can get accelerated failure models with these distributions using survreg() in the survival package. -thomas I do not need a accelerated failure model, but a proportional hazard model with a f0= weibull, exponential, loglogistic or lognormal baseline distribution. The hazard function is lambda(t)=exp(Xi*beta)*lambda0(t), where lambda0 is the baseline hazard lambda0(t)=f0(t)/(1-F0(t)) where f0 and F0 are the baseline density and cumulative distribution functions. This is a proportional hazard model since the ratio lambda(t|Xi)/lambda(t|Xj)=exp(Xi*beta)/exp(Xj*beta) does not depend on t. Valentin That will be a very strange model that I've never seen used before in survival analysis. Interpretation of parameters other than the hazard ratio may be tricky. Why do you need a nontraditional model such as this? Frank -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ 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] parametric proportional hazard regression
Valentin Dimitrov wrote: Cody, I have tried the survreg() in the Design library, it is analogous to survreg() in the survival library and it seems to me it is designed only for accelerated time models like the accelerated failure model (or accelerated lifetime model) and not for proportional hazard models. Correct me if I am wrong. Valentin survreg does not exist in the Design package. You may be thinking of psm which is a wrapper function for survreg. --- Hamilton, Cody [EMAIL PROTECTED] wrote: Valentin, Have you tried survreg() in the Design library? Regards, -Cody __ 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 -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ 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] heteroskedastic ordered probit
Dear all, is there existing code to fit a ordinal regression model with probit link where the variance can be modeled? For example, the polr function from the MASS package fits: probitP(Y = k | x) = zeta_k - eta Instead of, probitP(Y = k | x) = (zeta_k - eta)/scale, where scale = exp(b*z) which is what I need. I've seen these models go by the name of heteroskedastic ordered probit, location-scale glms, generalized nonlinear models, ... . I know the package gnlm (Jim Lindsey) has a function to fit this model using the logit link, so I'm wondering if this is already done in R using probit instead. Thanks. Regina __ 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