Re: [R] Rounding problem R vs Excel
On Tue, 3 Jun 2003 09:49:44 +0100, you wrote in message [EMAIL PROTECTED]: Duncan If the numbers are not represently exactly how does R resolve problems like the one below? Is there something that needs to be set up in the R environment like the number of significant figures? x-4.145*100+0.5 x [1] 415 floor(x) [1] 414 R doesn't do anything to resolve this problem; it's just the way the IEEE standard floating point formats work. In Excel 97, 4.145*100+0.5 is exactly equal to 415; I would guess this is either because they use a binary coded decimal format instead of the IEEE floating point types, or they round results internally in some way. R doesn't support BCD formats, and doesn't do tricky rounding behind your back. You get what you ask for. If you want the calculation above to give you exactly 415, the standard workaround in languages without BCD formats is to work in some decimal multiple of the actual numbers you're interested in, e.g. 1. Then you would store 4.145 as 41450, multiply by 100 (i.e. 100*1) and divide by 1 to give 4145000, and add 5000, to give 415. All of these numbers are exactly representable in double precision floating point types, because they are all integers with fewer than 53 bits in their binary representations. Doing this means you need to change the definitions of *, /, ^, and lots of other low level functions, but + and - work in the usual way. It might be an interesting project to write a package that does all of this. Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] (no subject)
Although not immediately relevant to the present inquiry this might still be an opportune moment to mention again an item that is at the top of my R-wish-listr: The SparseM function slm (for sparse lm) is well suited for problems like this in which the design matrix is quite sparse, so it would be great to have a version of model.matrix that would return a matrix in one of the formats of SparseM. I've hesitated to dig into this having looked a bit at the C, but if some brave soul were looking for a nice well-defined summer amusement... One can handle this by hand in specific instances, but it would be great to have a more automated way to do this via the formula approach. I think it is not uncommon in large regression problems that this would signficantly expand the range of applications that could be handled by R, especially on smaller machines. Regression/anova problems need only store the non-zero elements of X and computational effort on such problems should grow roughly proportionally to the number of these non-zero elements. Roger url:www.econ.uiuc.edu Roger Koenker Dept. of Economics UCL, email [EMAIL PROTECTED] Department of Economics Drayton House, vox:217-333-4558University of Illinois 30 Gordon St, fax:217-244-6678Champaign, IL 61820 London,WC1H 0AX, UK vox:020-7679-5838 On Tue, 3 Jun 2003, Ida Scheel wrote: It is over 3000 levels. I have enough RAM to do it, and I have run smaller examples (smaller datasets) with the same code which works. Peter Dalgaard BSA wrote: Ida Scheel [EMAIL PROTECTED] writes: Hei, I am trying to fit an ANOVA-model by lm. I get the error-message Error in lm.fit(x, y, offset = offset, ...) : negative length vectors are not allowed which I don't understand. My data looks fine, but one factor has extremely many levels. Does anyone have a tip? Not really. It sounds like it might be a bug. How many levels? Can you generate a simple example (possibly with simulated data) showing the same behaviour? [[alternate HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: Rtips (was Re: [R] ? building a database with a the great examples
I think there is a lot of merit in this idea. I think there is a big question about authentication and protection of Wikis from vandalism. I've set up Wikis for other projects that I started after Rtips. I have not seen your Wiki software before, but it looks pretty nice. I see it does have diff support, so old pages can be restored, yes? But it doesn't authenticate users, which causes me some concern. (I understand the Wiki philosophy that we should not be concerned about authentication, but I've never bought into it all the way). I have a TWiki site here: http://www.ku.edu/cgiwrap/pauljohn/twiki/view This one I hacked up special to use authentication on some pages so that people have to log in before they can edit. I had not realized before I looked at your page that Wiki implementations are customized for document format. For page sections, your Wiki uses = aHeading = but Twiki uses ---+ aHeading That's kindof a bummer. Detlef Steuer wrote: On 02-Jun-2003 Paul E. Johnson wrote: Perhaps you want to start maintaining Rtips itself! Perhaps it is time to start a wiki for R? For those not familiar with the idea of wikis look here: http://www.wikipedia.org/ (incredible wiki encyclopedia) http://www.wikipedia.org/wiki/WikiWiki (for a description of the mechanisms of wikiwikis) I just did a quick hack to set one up: http://fawn.unibw-hamburg.de/cgi-bin/Rwiki.pl?RwikiHome Any comments welcome! If the community uses the site I promise to do what I can to keep it running. Detlef -- Paul E. Johnson email: [EMAIL PROTECTED] Dept. of Political Sciencehttp://lark.cc.ukans.edu/~pauljohn University of Kansas Office: (785) 864-9086 Lawrence, Kansas 66045FAX: (785) 864-5700 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] S+ style implementation of GAM for R?
Hi, I've got the R library mgcv for GAM written by Simon Wood which works well in many instances. However, over the years I got attached to the S+ implementation of GAM which allows loess smoothing in more than 1 dimension as well as spline smoothing. Has anyone ported the S+ GAM library to R? Regards, Doug Beare. Fisheries Research Services, Marine Laboratory, Victoria Road, Torry, Aberdeen, UK. Tel. 44 (0) 1224 295314 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
RE: [R] tseries adf.test
I have a question regarding the adf.test command in the tseries library. I have a vector of time series observations (2265 daily log prices for the OEX to be exact). I also have this same data in first-differenced form. I want to test both vectors individually for staionarity with an Augmented Dickey-Fuller test. I noticed when I use the adf.test command from the tseries library, the general regression command used incorporates a constant and a linear trend -- (trend order of 1, I presume). My specific questions are as follows: (1) is it possible to alter the function to use a regression that does not incluse a linear trend? , because (2) it seems to me that I do not need to detrend if I've already taken first differences. Thanks in advance for your assistance. Rick Hello Rick, you might find the following link useful: http://www.econ.uiuc.edu/~econ472/tutorial9.html Pls note, that one typically follows a testing strategy in order to infer the characteristics of the time series in question (pure random walk, random with drift or random walk with drift and deterministic trend). The F-type test statistics (denoted by phi1, phi2, phi3 in the literature) can be calculated by making use of anova() and checking against the relevant critical values of these test statistics. HTH, Bernhard -- If you have received this e-mail in error or wish to read our e-mail disclaimer statement and monitoring policy, please refer to http://www.drkw.com/disc/email/ or contact the sender. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] S+ style implementation of GAM for R?
I've got the R library mgcv for GAM written by Simon Wood which works well in many instances. However, over the years I got attached to the S+ implementation of GAM which allows loess smoothing in more than 1 dimension as well as spline smoothing. Has anyone ported the S+ GAM library to R? - I've not come across a port (but note that mgcv does allow multidimensional smooths, albeit spline based and not loess). best, Simon _ Simon Wood [EMAIL PROTECTED]www.stats.gla.ac.uk/~simon/ Department of Statistics, University of Glasgow, Glasgow, G12 8QQ Direct telephone: (0)141 330 4530 Fax: (0)141 330 4814 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] libraries in .First
Dear all, I have a question regarding the .First function. I have included help.start() and several libraries in my .First as: .First - function() { help.start(browser = netscape7) library(lattice) library(modreg) library(splines) library(MASS) library(maps)} The libraries maps and splines do not seem to be available when I start R, but I have found if I change the order of the libraries in .First, these libraries may be available and others may not. Has anyone had a similar problem, or could anyone offer some suggestions? Thank you! Laurie Sindlinger __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] X11 not available
Hi, I just installed R in a Linux box. Unfortunately, I don't have root access and had to install it in my home directory. The software runs fine, but I cannot make it print graphics to the screen. If I type X11 I get: X11 is not available (though it is working for other software). I noticed that it prints graphs to a default ps file. Any suggestions about how to get around this? Many thanks, FG __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
RE: [R] Building an R package under Windows NT
The version of Perl I have is: This is perl, version 5.005_02 built for MSWin32-x86-object Copyright 1987-1998, Larry Wall Binary build 506 provided by ActiveState Tool Corp. http://www.ActiveState.com Built 15:40:37 Oct 27 1998 It was installed by my IS department some time ago. I will try installing Perl 5.8 and see if that works. Since Rcmd build -h does not print a description of the acceptable values for --docs=, I thought I would look for something similar. The --docs=TYPE option for build and the --type=TYPE for Rdconv looked related so I thought I would give it a try. Without knowing what the correct values for --docs=TYPE are, how am I suppose to know that --docs=TYPE is not the same as --type=TYPE for specifing what types of help documents to create? I don't think it is a stretch to see how these two could be confused. -Original Message- From: Prof Brian Ripley [mailto:[EMAIL PROTECTED] Sent: Tuesday, June 03, 2003 12:48 AM To: STABLER Benjamin Subject: RE: [R] Building an R package under Windows NT On Mon, 2 Jun 2003 [EMAIL PROTECTED] wrote: Professor Ripley, I just downloaded and installed the most current tools from your site (http://www.stats.ox.ac.uk/pub/Rtools/), I have version 5.005_02 of Perl, and I still get the same error. Do you think Perl 5.8 would *fix* this problem? I have no idea, but I do know that correcting all of *your* errors would solve the problem. As that version of Perl was AFAIK never released for Windows (and has not been available for several years if it was), I think you need to try to follow the instructions *exactly* (which you have not done re Perl, which says The Windows port of perl5, available via http://www.activestate.com/Products/ActivePerl/. BEWARE: you do need the *Windows* port and not the Cygwin one. ). I didn't think the docs=html option would work as I got the html option from the Rdconv type=TYPE help since the build help does not list the options. I didn't think to look at the install help. Why do you think --type (sic) takes the same values as --docs ? -- 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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] libraries in .First
If this is R 1.7.0, the preferred mechanism is options(defaultPackages), as in the following (from the examples in .First) # Example of Rprofile.site local({ old - getOption(defaultPackages) options(defaultPackages = c(old, MASS)) }) .First is run too early to be loading packages. Oh, and the preferred way to set the browser is environmental variable R_BROWSER, usually in R_HOME/etc/Renviron{.site}. On Tue, 3 Jun 2003, Laurie Sindlinger wrote: Dear all, I have a question regarding the .First function. I have included help.start() and several libraries in my .First as: .First - function() { help.start(browser = netscape7) library(lattice) library(modreg) library(splines) library(MASS) library(maps)} The libraries maps and splines do not seem to be available when I start R, but I have found if I change the order of the libraries in .First, these libraries may be available and others may not. Has anyone had a similar problem, or could anyone offer some suggestions? Thank you! Laurie Sindlinger __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help -- 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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
RE: [R] Building an R package under Windows NT
The emme2 package is not in the $RHOME/library directory. My sh.exe is the most current one from the tools available at http://www.stats.ox.ac.uk/pub/Rtools/.I searched for sh.exe and that is only one I've got on my system. -Original Message- From: Duncan Murdoch [mailto:[EMAIL PROTECTED] Sent: Monday, June 02, 2003 5:09 PM To: STABLER Benjamin Cc: [EMAIL PROTECTED] Subject: Re: [R] Building an R package under Windows NT On Mon, 2 Jun 2003 14:08:55 -0700 , you wrote: Thanks for the suggestions. 1) I fixed the zip.exe PATH issue. 2) I removed unnecessary quotes around C:\Program Files\R 3) I ran Rcmd install emme2 with the following result: Where is emme2? You shouldn't keep the source in the $RHOME/library/emme2 directory, which is where it will be installed. -- Making package emme2 adding build stamp to DESCRIPTION installing R files 175373 [main] sh 352 proc_subproc: Couldn't duplicate my handle0xA4 for pid 1736008448, Win32 error 6 This is an error in sh.exe. Are you getting the one from the R toolset, or possibly another one? Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] kmeans
Dear helpers I was working with kmeans from package mva and found some strange situations. When I run several times the kmeans algorithm with the same dataset I get the same partition. I simulated a little example with 6 observations and run kmeans giving the centers and making just one iteration. I expected that the algorithm just allocated the observations to the nearest center but think this is not the result that I get... Here are the simulated data dados-matrix(c(-1,0,2,2.5,7,9,0,3,0,6,1,4),6,2) dados [,1] [,2] [1,] -1.00 [2,] 0.03 [3,] 2.00 [4,] 2.56 [5,] 7.01 [6,] 9.04 plot(dados) dados-matrix(c(-1,0,2,2.5,7,9,0,5,0,6,1,4),6,2) plot(dados) A-kmeans(dados,dados[c(3,4),],1) A $cluster [1] 1 1 1 1 2 2 $centers [,1] [,2] 1 0.875 2.75 2 8.000 2.50 $withinss [1] 38.9375 6.5000 $size [1] 4 2 Any hints? Thanks a lot Luis Silva __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Update VR_7.1-6
The update of VR by downloading VR_7.1-6.zip and using install.packages (from local zip files) fails with the following error message: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `VR/DESCRIPTION' Other packages can be installed without problems, except of dse_2003.4-1 with a similar error message. Why? Operating System: Windows NT (4.0) R Version: R 1.7.0 * Dr. Dirk Enzmann Criminological Research Institute of Lower Saxony Luetzerodestr. 9 D-30161 Hannover Germany phone: +49-511-348.36.32 fax: +49-511-348.36.10 email: [EMAIL PROTECTED] http://www.kfn.de __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Update VR_7.1-6
On Tue, 3 Jun 2003, Dirk Enzmann wrote: The update of VR by downloading VR_7.1-6.zip and using install.packages (from local zip files) fails with the following error message: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `VR/DESCRIPTION' Other packages can be installed without problems, except of dse_2003.4-1 with a similar error message. Because there is a bug in the handling of package bundles in R 1.7.0. VR and dse are bundles. R1.7.1 will be coming out soon and fixes this. Or you can manually unzip the bundle in the appropriate place in the library directory. -thomas __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Update VR_7.1-6
Hi Dirk, dse and VR are bundles or packages. There is a bug in rw1070 in the way it installs bundles. Download the zip file directly and unzip it to the library directory in wherever R is installed on your system. This has been discussed recently (this morning) on the list. Search the archives to read more. G Dirk Enzmann wrote: The update of VR by downloading VR_7.1-6.zip and using install.packages (from local zip files) fails with the following error message: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `VR/DESCRIPTION' Other packages can be installed without problems, except of dse_2003.4-1 with a similar error message. Why? Operating System: Windows NT (4.0) R Version: R 1.7.0 * Dr. Dirk Enzmann Criminological Research Institute of Lower Saxony Luetzerodestr. 9 D-30161 Hannover Germany phone: +49-511-348.36.32 fax: +49-511-348.36.10 email: [EMAIL PROTECTED] http://www.kfn.de __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [T] +44 (0)20 7679 5522 ENSIS Research Fellow [F] +44 (0)20 7679 7565 ENSIS Ltd. ECRC [E] [EMAIL PROTECTED] UCL Department of Geography [W] http://www.ucl.ac.uk/~ucfagls/cv/ 26 Bedford Way[W] http://www.ucl.ac.uk/~ucfagls/ London. WC1H 0AP. %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Update VR_7.1-6
I have explained that on R-help earlier today (and when VR_7.1-4 was announced). Please look back a few hours in the R-help postings. On Tue, 3 Jun 2003, Dirk Enzmann wrote: The update of VR by downloading VR_7.1-6.zip and using install.packages (from local zip files) fails with the following error message: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `VR/DESCRIPTION' Other packages can be installed without problems, except of dse_2003.4-1 with a similar error message. Why? Because VR and dse are bundles and someone broke the installation of bundles in R 1.7.0 on Windows. You can download and unzip the zip files in rw1070/library. -- 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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] kmeans
On Tue, 3 Jun 2003, Luis Miguel Almeida da Silva wrote: I was working with kmeans from package mva and found some strange situations. When I run several times the kmeans algorithm with the same dataset I get the same partition. Why does that surprise you? I simulated a little example with 6 observations and run kmeans giving the centers and making just one iteration. I expected that the algorithm just allocated the observations to the nearest center but think this is not the result that I get... That's not what the documentation says it does: The data given by `x' is clustered by the k-means algorithm. When this terminates, all cluster centres are at the mean of their Voronoi sets (the set of data points which are nearest to the cluster centre). which is true in your example. It has run one iteration of re-allocation; as you can see by reading the source code or the reference. [...] -- 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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] coefficient of logistic regression
Ahmet - In a logistic regression model, fitted probabilities make sense for individual cases (rows in the data set), as well as for future cases (predictions) for which no outcome (success or failure) has been observed yet. Fitted probabilities are calculated from the matrix formula: Pr[success] = exp( X %*% beta) / (1 + exp( X %*% beta) where X is an [n x (p+1)] matrix, containing all p predictor variables as columns, preceded by a column of 1s for the intercept, and beta is the [(p+1) x 1] vector of logistic regression coefficients. One can interpret the sign and the magnitude of an individual regression coeffient by saying that an increase of 1 unit in predictor variable [i] will increase or decrease the odds of success by a multiplier of exp(beta[i]). When beta[i] 0 the odds increase, because exp(beta[i]) 1, and when beta[i] 0 the odds decrease, because exp(beta[i]) 1. I hope this explanation helps. - tom blackwell - u michigan medical school - ann arbor - On Tue, 3 Jun 2003, orkun wrote: Hello in logistic regression, I want to know that it is possible to get probability values of each predictors by using following formula for each predictor one by one (keeping constant the others) exp(coef)/(1+exp(coef)) thanks in advance Ahmet Temiz __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] rw1070 and package bundles; was: Update VR_7.1-6
Prof Brian Ripley wrote: I have explained that on R-help earlier today (and when VR_7.1-4 was announced). Please look back a few hours in the R-help postings. On Tue, 3 Jun 2003, Dirk Enzmann wrote: The update of VR by downloading VR_7.1-6.zip and using install.packages (from local zip files) fails with the following error message: Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `VR/DESCRIPTION' Other packages can be installed without problems, except of dse_2003.4-1 with a similar error message. Why? Because VR and dse are bundles and someone broke the installation of bundles in R 1.7.0 on Windows. You can download and unzip the zip files in rw1070/library. The CRAN maintainers also got first messages from users who didn't know about this bug. In order to avoid a huge amount of further questions and bug reports related to these problems, I decided to move the two package bundles from CRAN/bin/windows/contrib/1.7 to a subfolder until R-1.7.1 will be released. Please read CRAN/bin/windows/contrib/1.7/ReadMe for details. The change presumably will show up on CRAN master within 24 hours. Uwe Ligges __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
RE: [R] Building an R package under Windows NT
I finally figured out the problem. I went ahead and installed Perl 5.8 but that didn't do it. The problem was that an existing program's directory was earlier in the PATH and so one (or more) of the components of the Rcmd INSTALL was an older version. It wasn't zip since I renamed the old zip. Anyway, thanks for your help. Regards, Ben Stabler -Original Message- From: Prof Brian Ripley [mailto:[EMAIL PROTECTED] Sent: Tuesday, June 03, 2003 12:48 AM To: STABLER Benjamin Subject: RE: [R] Building an R package under Windows NT On Mon, 2 Jun 2003 [EMAIL PROTECTED] wrote: Professor Ripley, I just downloaded and installed the most current tools from your site (http://www.stats.ox.ac.uk/pub/Rtools/), I have version 5.005_02 of Perl, and I still get the same error. Do you think Perl 5.8 would *fix* this problem? I have no idea, but I do know that correcting all of *your* errors would solve the problem. As that version of Perl was AFAIK never released for Windows (and has not been available for several years if it was), I think you need to try to follow the instructions *exactly* (which you have not done re Perl, which says The Windows port of perl5, available via http://www.activestate.com/Products/ActivePerl/. BEWARE: you do need the *Windows* port and not the Cygwin one. ). I didn't think the docs=html option would work as I got the html option from the Rdconv type=TYPE help since the build help does not list the options. I didn't think to look at the install help. Why do you think --type (sic) takes the same values as --docs ? -- 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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] (no subject)
Hi, I would like to know if it is possible to get printed output while a loop is taking place. Example: for(i in 1:10){ print(i) some long process } This will print the values of i only after the loop is finished, what I would like is to see them when the process enters the i-th iteration to keep track of how the program is running. Thank you, Gilda __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] (no subject)
Gilda Garibotti [EMAIL PROTECTED] writes: Hi, I would like to know if it is possible to get printed output while a loop is taking place. Example: for(i in 1:10){ print(i) some long process } This will print the values of i only after the loop is finished, what I would like is to see them when the process enters the i-th iteration to keep track of how the program is running. Windows, right? (This is system dependent) There's a menu item entitled Buffer output or something to that effect. Turn it off and print() calls display immediately. Lengthy output becomes slower, though. -- O__ Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] coefficient of logistic regression
Dear Ahmet, Sorry for the slow response, but I've been busy all today, coincidentally teaching a workshop on logistic regression. Tom Blackwell sent you a useful suggestion for interpreting coefficients on the odds scale. If you want to trace out the partial relationship of the fitted probability of response to a particular predictor holding others constant, you can set the other predictors to typical values and let the predictor in question vary over its range, transforming the fitted log-odds to the probability scale. You may be interested in my effects package (on CRAN or at http://socserv.socsci.mcmaster.ca/jfox/Misc/effects/index.html), which makes these kinds of displays for linear and generalized-linear models, including those with interactions. Regards, John At 03:06 PM 6/3/2003 +0300, orkun wrote: John Fox wrote: At 11:54 AM 6/3/2003 +0300, orkun wrote: in logistic regression, I want to know that it is possible to get probability values of each predictors by using following formula for each predictor one by one (keeping constant the others) exp(coef)/(1+exp(coef)) Dear Ahmet, This will almost surely give you nonsense, since it produces a fitted probability ignoring the constant in the model (assuming that there is one), setting other predictors to 0 and the predictor in question to 1. What is it that you want to do? I hope that this helps, John thank you Say, I just want to find each predictor's particular effect on dependent variables. Actual model is to prepare landslide susceptibility map on GIS. So I want to know what the effect as probability value comes from each predictor. For instane what is the effect of slope on landslide susceptibility. Should I keep others constant ? kind regards - John Fox Department of Sociology McMaster University Hamilton, Ontario, Canada L8S 4M4 email: [EMAIL PROTECTED] phone: 905-525-9140x23604 web: www.socsci.mcmaster.ca/jfox __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Question about looking up names
I think I now understand how R looks up names. Could anyone tell me if I have this right? First it looks up the nested environments created by lexical scoping. Then, if it gets to the top (.GlobalEnv) it also looks through the list of things that have been attached. It never looks in the call stack unless you explicitly ask it to, or mess with the environment frames. The reason I ask is that it's not entirely clear to me from the R Language Definition how these 3 search spaces (environments/lexical scoping; call stack/dynamic scoping; attach/search list) are related. For example the discussion of 3.5.3 (the call stack) observes that dynamic scoping contradicts the default scoping rules in R. I spent some time trying to figure out how it could do both, before deciding it doesn't. I suppose the implicit corollary of the contradiction referred to in 3.5.3--so we don't do that and you must intervene to achieve dynamic scoping--was obvious to the authors. It just wasn't obvious to me. Since I'm still not sure, I thought I'd check. Thanks. -- Ross Boylan wk: (415) 502-4031 530 Parnassus Avenue (Library) rm 115-4 [EMAIL PROTECTED] Dept of Epidemiology and Biostatistics hm: (415) 550-1062 University of California, San Francisco San Francisco, CA 94143-0840 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Question about looking up names
On 3 Jun 2003, Ross Boylan wrote: I think I now understand how R looks up names. Could anyone tell me if I have this right? First it looks up the nested environments created by lexical scoping. Then, if it gets to the top (.GlobalEnv) it also looks through the list of things that have been attached. It never looks in the call stack unless you explicitly ask it to, or mess with the environment frames. The reason I ask is that it's not entirely clear to me from the R Language Definition how these 3 search spaces (environments/lexical scoping; call stack/dynamic scoping; attach/search list) are related. For example the discussion of 3.5.3 (the call stack) observes that dynamic scoping contradicts the default scoping rules in R. I spent some time trying to figure out how it could do both, before deciding it doesn't. I suppose the implicit corollary of the contradiction referred to in 3.5.3--so we don't do that and you must intervene to achieve dynamic scoping--was obvious to the authors. It just wasn't obvious to me. Since I'm still not sure, I thought I'd check. Yes. -thomas __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Logistic regression problem: propensity score matching
Hello all. I am doing one part of an evaluation of a mandatory welfare-to-work programme in the UK. As with all evaluations, the problem is to determine what would have happened if the initiative had not taken place. In our case, we have a number of pilot areas and no possibility of random assignment. Therefore we have been given control areas. My problem is to select for survey individuals in the control areas who match as closely as possible the randomly selected sample of action area participants. As I understand the methodology, the procedure is to run a logistic regression to determine the odds of a case being in the sample, across both action and control areas, and then choose for control sample the control area individual whose odds of being in the sample are closest to an actual sample member. So far, I have following the multinomial logistic regression example in Fox's Companion to Applied Regression. Firstly, I would like to know if the predict() is producing odds ratios (or probabilities) for being in the sample, which is what I am aiming for. Secondly, how do I get rownames (my unique identifier) into the output from predict() - my input may be faulty somehow and the wrong rownames being picked up - as I need to export back to database to sort and match in names, addresses and phone numbers for my selected samples. My code is as follows: londonpsm - sqlFetch(channel, London_NW_london_pilots_elig, rownames=ORCID) attach(londonpsm) mod.multinom - multinom(sample ~ AGE + DISABLED + GENDER + ETHCODE + NDYPTOT + NDLTUTOT + LOPTYPE) lonoutput - predict(mod.multinom, sample, type='probs') london2 - data.frame(lonoutput) The Logistic regression seems to work, although summary() says the it is not a matrix. The output looks like odds ratios, but I would like to know whether this is so. Thank you Paul Bivand __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Question about looking up names
On Tue, 2003-06-03 at 16:34, Robert Gentleman wrote: Also, note that you can get the effect of lexical scope by doing things like Do you mean you can get the effects of dynamic scope? f- function(x) x+y e1 - new.env() assign(y, 10, env=e1) environment(f) - e1 #now like lexical scope; you can futz with f's environment, assigning, # modifying as you like P.S. Thanks to everyone who responded. So fast! __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Rounding problem R vs Excel
On Tue, 2003-06-03 at 08:32, Duncan Murdoch wrote: On Tue, 3 Jun 2003 09:49:44 +0100, you wrote in message [EMAIL PROTECTED]: Duncan If the numbers are not represently exactly how does R resolve problems like the one below? Is there something that needs to be set up in the R environment like the number of significant figures? x-4.145*100+0.5 x [1] 415 floor(x) [1] 414 R doesn't do anything to resolve this problem; it's just the way the IEEE standard floating point formats work. In Excel 97, 4.145*100+0.5 is exactly equal to 415; I would guess this is either because they use a binary coded decimal format instead of the IEEE floating point types, or they round results internally in some way. R doesn't support BCD formats, and doesn't do tricky rounding behind your back. You get what you ask for. If you want the calculation above to give you exactly 415, the standard workaround in languages without BCD formats is to work in some decimal multiple of the actual numbers you're interested in, e.g. 1. Then you would store 4.145 as 41450, multiply by 100 (i.e. 100*1) and divide by 1 to give 4145000, and add 5000, to give 415. All of these numbers are exactly representable in double precision floating point types, because they are all integers with fewer than 53 bits in their binary representations. Doing this means you need to change the definitions of *, /, ^, and lots of other low level functions, but + and - work in the usual way. It might be an interesting project to write a package that does all of this. Duncan Murdoch In Excel, the IEEE standard (754) is used to internally represent floats. A MS-KB article on this is here: http://support.microsoft.com/default.aspx?scid=kb;[LN];214118 Another, more detailed, is here: http://support.microsoft.com/default.aspx?scid=kb;EN-US;78113 What is curious about this situation, and apropos to Prof. Ripley's comments about the difference between internal representation, rounding and displayed values, is the following information. Note how the results of cell calculations differ between Excel, OpenOffice.org Calc and Gnumeric. In each case, I use a format setting of 20 digits after the decimal place with scientific notation. This is best read with a fixed width font. OOo Calc 1.0.2 and 1.1 Beta2: Cell Formula Value = 4.145 * 100 + 0.5 4.1500E+02 = 0.5 - 0.4 - 0.1 0.E+00 =(0.5 - 0.4 - 0.1)0.E+00 Excel 2002 (XP): Cell Formula Value = 4.145 * 100 + 0.5 4.1500E+02 = 0.5 - 0.4 - 0.1 0.E+00 =(0.5 - 0.4 - 0.1)-2.7755575615628900E-17 Gnumeric 1.0.12: Cell Formula Value = 4.145 * 100 + 0.5 +4.14943157E+02 = 0.5 - 0.4 - 0.1 -2.77555756156289135106E-17 *Gnumeric does not appear to allow the surrounding parens. For comparison, R 1.7.1 Beta under RH 9 and WinXP: print(4.145 * 100 + 0.5, digits = 20) [1] 414.94 formatC(4.145 * 100 + 0.5, format = E, digits = 20) [1] 4.14943157E+02 print(0.5 - 0.4 - 0.1, digits = 20) [1] -2.775557561562891e-17 formatC(0.5 - 0.4 - 0.1, format = E, digits = 20) [1] -2.77555756156289135106E-17 What is interesting is the change in the displayed value in Excel when the second formula is surrounded by parens (which I found purely by accident). This would suggest that there may be something going on in the parsing of the cell formula that affects the calculation and displayed value. Also note the precision of the resultant number. Presuming that each of the spreadsheet programs are using IEEE standard internal representation, there are clearly differences in the way in which each visually displays the values, both by default and when explicitly formatted. Using the following cell formula: = 1.333 + 1.225 - 1.333 - 1.225 there is an indication in the second MS-KB article above, that Excel 97 introduced an optimization dealing with results near zero. The example above when performed in Excel 97 and later correctly displays 0 or 0.000E+00 in scientific notation. whereas Rather than displaying 0, Excel 95 displays -2.22044604925031E-16. The terms optimization and correctly displays are an interesting choice of words. I have a post to one of the OOo forums regarding my inability to replicate the IEEE precision issues in Calc under any circumstances using the three formulas and any numeric formatting options. It may be that the OOo folks copied the MS Excel optimization with no override. FYI...the IEEE has a reference site for the standard here: http://grouper.ieee.org/groups/754/ HTH, Marc Schwartz __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] X11 not available
Hello, | From: [EMAIL PROTECTED] | Date: Tue, 03 Jun 2003 16:23:20 + | | Hi, | | I just installed R in a Linux box. Unfortunately, I don't have root access and | had to install it in my home directory. | The software runs fine, but I cannot make it print graphics to the screen. If I | type X11 I get: X11 is not available (though it is working for other software). | I noticed that it prints graphs to a default ps file. | Any suggestions about how to get around this? | | Many thanks, | | FG perhaps you have compiled it without x11 support. The reason could be missing XFree86-devel.rpm or something similar. Have you checked whad did the configuration script say? Just a suggestion Ott __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Your Message Contained a Potential Virus
[EMAIL PROTECTED] writes: Result: Virus Detected Virus Name: [EMAIL PROTECTED] File Attachment: screensaver.scr Attachment Status: deleted --- Original message information follows --- From: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Date: Tue, 3 Jun 2003 21:38:00 --0400 Subject: Re: Application Received: (from HZHANG2-LAP [24.74.137.211]) by mskavs1.mskcc.org (SAVSMTP 3.1.1.32) with SMTP id M2003060321303704373 for [EMAIL PROTECTED]; Tue, 03 Jun 2003 21:30:38 -0400 Grrr Well they should go talk to mr./ms. Zhang about the virus on his/her laptop shouldn't they? In any case, as far as I understand the mail standards, delivery errors should go to the contents of the Sender: field, not From:, exactly to avoid bothering entire mailing lists. -- O__ Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] plot rpart tree's from list object
Hello, i want the post plot's from a rpart list object with 18 tree's , getting no error - but getting no files,too? Perhaps i should using assign!? for (i in 1:length(treeList)) { post(treeList[[i]],filename=paste(Tree,i,sep=.ps), title=Arbeitszufriedenheit, digits=getOption(digits) - 0,use.n=TRUE) } many thanks for help, christian [[alternate HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] (no subject)
Hoi Peter, --On woensdag 4 juni 2003 0:16 +0200 Peter Dalgaard BSA [EMAIL PROTECTED] wrote: Gilda Garibotti [EMAIL PROTECTED] writes: Hi, I would like to know if it is possible to get printed output while a loop is taking place. Example: for(i in 1:10){ print(i) some long process } This will print the values of i only after the loop is finished, what I would like is to see them when the process enters the i-th iteration to keep track of how the program is running. Windows, right? (This is system dependent) There's a menu item entitled Buffer output or something to that effect. Turn it off and print() calls display immediately. Lengthy output becomes slower, though. If you don't want to depend on you (or other people) turning of the buffering, use something like cat(this or that); flush.console. regards, Paul -- Paul Lemmens NICI, University of Nijmegen ASCII Ribbon Campaign /\ Montessorilaan 3 (B.01.03)Against HTML Mail \ / NL-6525 HR Nijmegen X The Netherlands / \ Phonenumber+31-24-3612648 Fax+31-24-3616066 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Logistic regression problem: propensity score matching
1) Why are you using multinom when this is not a multinomial logistic regression? You could just use a binomial glm. 2) The second argument to predict() is `newdata'. `sample' is an R function, so what did you mean to have there? I think the predictions should be a named vector if `sample' is a data frame. 3) There are many more examples of such things (and more explanation) in Venables Ripley's MASS (the book). On Wed, 4 Jun 2003, Paul Bivand wrote: I am doing one part of an evaluation of a mandatory welfare-to-work programme in the UK. As with all evaluations, the problem is to determine what would have happened if the initiative had not taken place. In our case, we have a number of pilot areas and no possibility of random assignment. Therefore we have been given control areas. My problem is to select for survey individuals in the control areas who match as closely as possible the randomly selected sample of action area participants. As I understand the methodology, the procedure is to run a logistic regression to determine the odds of a case being in the sample, across both action and control areas, and then choose for control sample the control area individual whose odds of being in the sample are closest to an actual sample member. So far, I have following the multinomial logistic regression example in Fox's Companion to Applied Regression. Firstly, I would like to know if the predict() is producing odds ratios (or probabilities) for being in the sample, which is what I am aiming for. You asked for `probs', so you got probabilities. Secondly, how do I get rownames (my unique identifier) into the output from predict() - my input may be faulty somehow and the wrong rownames being picked up - as I need to export back to database to sort and match in names, addresses and phone numbers for my selected samples. My code is as follows: londonpsm - sqlFetch(channel, London_NW_london_pilots_elig, rownames=ORCID) attach(londonpsm) mod.multinom - multinom(sample ~ AGE + DISABLED + GENDER + ETHCODE + NDYPTOT + NDLTUTOT + LOPTYPE) lonoutput - predict(mod.multinom, sample, type='probs') london2 - data.frame(lonoutput) The Logistic regression seems to work, although summary() says the it is not a matrix. what is `it'? The output looks like odds ratios, but I would like to know whether this is so. No. -- 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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: Rtips (was Re: [R] ? building a database with a the great examples
On 03-Jun-2003 Paul E. Johnson wrote: I think there is a lot of merit in this idea. I think there is a big question about authentication and protection of Wikis from vandalism. Yes. But I`ll do my daily backups, so major vandalism wouldn't be such a problem. (I hope.) Personally I don't like to have a list of people who are allowed to edit the pages. (and I dont like to keep such a list current ...) Under preferences you can choose to give yourself an identity, but that's not for authentication. I've set up Wikis for other projects that I started after Rtips. I have not seen your Wiki software before, but it looks pretty nice. I see it does have diff support, so old pages can be restored, yes? Yes! It keeps the diffs for quite some time. So if anyone realizes minor, i.e. pagewise, vandalism content can be rebuild. But it doesn't authenticate users, which causes me some concern. (I understand the Wiki philosophy that we should not be concerned about authentication, but I've never bought into it all the way). I think the R community is growing very fast. The work to give passwords to interested people frightens me. And I would never ask for such a password personally. I'm willing to write some short note _now_, but not if I have to wait a day or so to get access. Authentication fits if you have a well defined group you expect to add contents. My position therefore: give it a try. If vandalism turns out to be a problem, I'll have to think about it. I have a TWiki site here: http://www.ku.edu/cgiwrap/pauljohn/twiki/view This one I hacked up special to use authentication on some pages so that people have to log in before they can edit. I had not realized before I looked at your page that Wiki implementations are customized for document format. For page sections, your Wiki uses = aHeading = but Twiki uses ---+ aHeading That's kindof a bummer. Yes. I don't understand the authors of wikis, too. I just chose a recommended one. If there is anything serious against UseModWiki, _now_ would be the time to switch. UseMod is very easy to set up, an needs little resources. These two points are very important for me. detlef __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Your Message Contained a Potential Virus
Text: This is an automated message. Please read it carefully. You should know that your recent email message detailed below, to Memorial Sloan-Kettering Cancer Center, was identified as potentially containing a virus. From: [EMAIL PROTECTED] To: [EMAIL PROTECTED] If we have been able to repair your message, it has been delivered. If we have not, it has been blocked and the recipient(s) of your email have been informed that the email has been blocked. Please note that as a matter of policy certain attachment types are blocked by default based on the file extension. If you have a legitimate file to send which is being blocked, please consider renaming the extension or enclosing it within a ZIP file. Thank You Viruses found: --- Scan information follows --- Result: Virus Detected Virus Name: [EMAIL PROTECTED] File Attachment: screensaver.scr Attachment Status: deleted --- Original message information follows --- From: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Date: Tue, 3 Jun 2003 21:38:00 --0400 Subject: Re: Application Received: (from HZHANG2-LAP [24.74.137.211]) by mskavs1.mskcc.org (SAVSMTP 3.1.1.32) with SMTP id M2003060321303704373 for [EMAIL PROTECTED]; Tue, 03 Jun 2003 21:30:38 -0400 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Mode of MCMC chain
Hello, are there any functions in R for estimation of the mode of a MCMC-chain? Best, Patrik Waldmann [[alternate HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Strip location and grid colour in Lattice
I am probably missing something quite obvious, but any help would be appreciated. I am continually getting people misreading the lattice plots because they are expecting the strip (with the factor names in them) to be below the graph. Is there anyway of achieving this. Secondly, from a more personal note I find the grid formed by the axes to be a bit overpowering and would like to make it a little less bold by changing it to a grey of some kind. I can't see that the scales options have anythig in their that I could use. I can change the label colours and tick marks, but then I draw a blank. While I'm on a role, I find that quite often I have to resort to the at and label sections of the scales function to get my tickmarks looking OK. This seems to be when am producing line graphs with one of the scales being a date (POSIXct). What is not clear to me is if all POSIXct variables are the same. The xyplot doco indicates that the at co-ordinates should be native co-ordinates. Can anyone point me to where in the voluminous documentation one looks to understand what this means. I have found that on some occasions the co-ordinates are in seconds (as the documentation on POSIXct states, but this afternoon I found that the values seemed to be in years. Which wasn't a problem other than I wish I could understand what was actually happening. For the years example, when the data is originally imported the years came in as integers. str(rbd) `data.frame': 541 obs. of 6 variables: $ Year: int 1993 1994 1995 1996 1997 1998 1999 2000 2001 1993 ... $ Hosp: Factor w/ 75 levels ALBANY HOSP..,..: 23 23 23 23 23 23 23 23 23 28 ... $ Beddays : int 2431 2507 2201 2985 2702 2461 2535 2970 3271 1246 ... $ HD : Factor w/ 21 levels Avon HD.,Bunb..,..: 10 10 10 10 10 10 10 10 10 10 ... $ HR : Factor w/ 6 levels Goldfields-..,..: 3 3 3 3 3 3 3 3 3 3 ... $ HospCode: int 127 127 127 127 127 127 127 127 127 128 ... Thinking that I needed a date I promptly put rbd$Year - as.POSIXct(ISOdate(rbd$Year,6,30)) then onwards and forwards for (h in levels(rbd$HR)){ HRData - subset(rbd,HR==h) HRData$CommnDesc - HRData$CommnDesc[,drop=T] temp -c((FormatLabels(levels(HRData$CommnDesc)[1],20))) for (j in 2:length(levels(HRData$CommnDesc))){ temp - c(temp,FormatLabels(levels(HRData$CommnDesc)[j],20)) } levels(HRData$CommnDesc) - temp p1 - bwplot(Beddays~Year |CommnDesc,HRData, panel = panel.linejoin, horizontal=F,bty=n, as.table=T, par.strip.text=list(lines=3.5,cex=0.8,style=1), main=paste(h,Inpatient Beddays), scales=list(x=list(cex=0.8,rot=90, at=c(2,6,9), labels=c(94,98,01),col=navy)) ) print(p1) savePlot(file=paste(OutputPath,Inpatient beddays -(lattice) by region,h, ,j,sep=),type=wmf) } Of course there are a few things in here that are probably not the right way to do things, but I tend to be more interested in the output, rather than whether or not my programming is up to speed. But it has been a little bug bear of mine about dropping factors when subsetting the data. I've noticed subset options as I've been going through assorted bits and pieces, but there never seems to be enough time to follow up. This is in striking contrast to a previous attempt (most of the code however is at home not here), but the functions that I worked out for the at and label functions were ProcLab - function(DateData,breakNum){ maxplot - round(as.numeric(max(DateData)),digits=0) minplot - round(as.numeric(min(DateData)),digits=0) maxplotnum- round(((maxplot-minplot)/86400)+1,digits=0) jumpnum - (maxplotnum/((breakNum)-1))*.98 lablist - seq(min(DateData),max(DateData),jumpnum*86400) } ProcAt - function(DateData,breakNum){ maxplot - round(as.numeric(max(DateData)),digits=0) minplot - round(as.numeric(min(DateData)),digits=0) maxplotnum- round(((maxplot-minplot)/86400)+1,digits=0) jumpnum - (maxplotnum/((breakNum)-1))*.98 atlist - seq(0,maxplotnum,jumpnum) } The kludges were in because without them the whole thing fell over, presumeably because I would needed to have set the limits as well. _ Tom Mulholland Senior Policy Officer WA Country Health Service 189 Royal St, East Perth, WA, 6004 Tel: (08) 9222 4062 e-mail: [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] The contents of this e-mail transmission are confidential and may be protected by professional privilege. The contents are intended only for the named recipients of this e-mail. If you are not the intended recipient, you are hereby notified that any use, reproduction, disclosure or distribution of the information contained in this e-mail is prohibited. Please notify the sender immediately. [[alternate HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] (no subject)
Hi everybody, I finally hope to reach the person who started the thread [R] (no subject). The innermost level of text was my original request. Kind regards Christian At 10:50 2003-06-04 +0200, you wrote: Hoi Christian, --On woensdag 4 juni 2003 10:36 +0200 Christian Hoffmann [EMAIL PROTECTED] wrote: Hi Paul At 08:44 2003-06-04 +0200, you wrote: Hoi Christian, --On woensdag 4 juni 2003 8:39 +0200 Christian Hoffmann [EMAIL PROTECTED] wrote: Please avoid no subject. There might be people (like me, when I am in a hurry or in a bad mood) who just delete such messages. It would be a pity to miss interesting information, yet. Please avoid prematurely emailing somebody who did not originate a thread that is under your current scrutiny! No offence meant, but... How could I find out this person if I already deleted so many messages of that kind? No offence either, but why mail me, and not the person who sent you the first mail (in this current thread i.e.) without the subject? IMHO either be consistent and mail everybody, or mail nobody!? regards, Paul -- Paul Lemmens NICI, University of Nijmegen ASCII Ribbon Campaign /\ Montessorilaan 3 (B.01.03)Against HTML Mail \ / NL-6525 HR Nijmegen X The Netherlands / \ Phonenumber+31-24-3612648 Fax+31-24-3616066 Dr.sc.math.Christian W. Hoffmann Mathematics and Statistical Computing Landscape Dynamics and Spatial Development Swiss Federal Research Institute WSL Zuercherstrasse 111 CH-8903 Birmensdorf, Switzerland phone: ++41-1-739 22 77fax: ++41-1-739 22 15 e-mail: [EMAIL PROTECTED] www: http://www.wsl.ch/staff/christian.hoffmann/ - Coordinator of cooperation WSL - UFU Ekaterinburg/Russia - Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html Dangers of .doc, .rtf: http://www.heise.de/newsticker/data/jk-27.01.02-001/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] predict.glm(glm.ob,type=terms)
hello pgeo-predict.glm(glm.ob,type=resp) works fine. But I need to get predictions values in terms of each factor variables. pgeo-predict.glm(glm.ob,type=terms) gives Error in rep(1/n,n) %*% model.matrix(object): non conformable arguments Could anyone tell me why ? Ahmet Temiz Turkey __ __ The views and opinions expressed in this e-mail message are the ... {{dropped}} __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] convert factor to numeric
Hi R-experts! Every once in a while I need to convert a factor to a vector of numeric values. as.numeric(myfactor) of course returns a nice numeric vector of the indexes of the levels which is usually not what I had in mind: v - c(25, 3.78, 16.5, 37, 109) f - factor(v) f [1] 25 3.78 16.5 37 109 Levels: 3.78 16.5 25 37 109 as.numeric(f) [1] 3 1 2 4 5 What I really want is a function unfactor that returns v: unfactor(f) [1] 25.00 3.78 16.50 37.00 109.00 Of course I could use something like as.numeric(levels(f)[as.integer(f)]) But I just can't believe there is no R function to do this in a more readable way. Actually, the behaviour of as.numeric() doesn't strike me as very intuitive. I'm sure it has been implemented that way for a reason - but what is it? cu Philipp -- Dr. Philipp PagelTel. +49-89-3187-3675 Institute for Bioinformatics / MIPS Fax. +49-89-3187-3585 GSF - National Research Center for Environment and Health Ingolstaedter Landstrasse 1 85764 Neuherberg Germany __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Rounding problem R vs Excel
On 04 Jun 2003 00:24:08 -0500, you wrote: Excel 2002 (XP): Cell Formula Value = 0.5 - 0.4 - 0.1 0.E+00 =(0.5 - 0.4 - 0.1)-2.7755575615628900E-17 ... What is interesting is the change in the displayed value in Excel when the second formula is surrounded by parens (which I found purely by accident). This would suggest that there may be something going on in the parsing of the cell formula that affects the calculation and displayed value. Interesting? I'd say horrifying. When (expr) does not evaluate the same as expr, what can you trust? Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help