similar to: (no subject)

Dear list, I calculated the kappa-accuracy for two differnt classifications. How can I test now the two kappa-value for significance? thanks, Mark ..................................................................... Markus Schwarz Wissenschaftliche Mitarbeiterin Eidg. Forschungsanstalt WSL Forschungsprogramm Musterland Z??rcherstrasse 111 CH-8903 Birmensdorf Telefon +41-44-739 22 87 Fax

Hi list I'm looking for statisitic measurements describing the pattern of points within a given polygon. Is there a function calculating the Index of Dispersion and/or are there other functions summarising an observed pattern? Markus Schwarz ..................................................................... Markus Schwarz Wissenschaftliche Mitarbeiterin Eidg. Forschungsanstalt WSL

Hello, I have two classifications. How can I compare the overall-accuracy of these classifications to each other? Is there a possibility within R to test if the achieved overall-accuracy for the classifications are differing significantly? Additionaly, are the McNemar-Test and Broker-Test implented in a package of R? Thank in advance for your help, Markus

Hi there, I am interested in the inner workings of wilcox.test: > wilcox.test function (x, ...) UseMethod("wilcox.test") <environment: namespace:stats> how can I get at the code, if it is R-code? For Methods one should be able to learn what extension to use, but here default or such do not help. Is there a wayplot.default to learn which different versions of

Hi, Sorry to bother, but I checked around and did not succed creating a bundle from six existing packages (which are checkable, installable, etc. individually). I carefully followed the procedure given in ch. 1.1.5 Package bundles. However, I am getting hoffmann at fluke:~/R/Sources >R CMD check cwhmisc * checking for working latex ... OK * using log directory

Hi, Using help under help.start() is working fine (html) Q1: using > ?solve results in the message "sh: ~/R/tmp/R4f32S1cfb: cannot open" displayed in the XEmacs buffer "help[R](solve)". BUT there exists a file R.R4f32S167e in that directory. (not quite the same name). Does the "sh: " prefix suggest that the shell "sh" is run, although my

Hi there, (sorry Martin, ich habe den falschen Empfaenger erwischt :-( ) Is there a function in the base package or elsewhere which returns TRUE and not NA, even if NA is involved: (I did not find in the help system :-( ) if ( x==something & !is.na(something)) or if ( x==something & !is.na(x==something)) If "something" is complicated, a function would be handy. Thanks

Hi there I hope that I am in the right forum. I am working on Win2000 PC connected via Exceed 6.0.1.0 to a SunOS fluke 5.9 Generic_118558-11 sun4u sparc SUNW,Sun-Fire-480R There I am using Mozilla Firefox 1.0.7 as a browser. I am having difficulties viewing .html files. Their first pages are displayed normally, black on white, links in blue. When I scoll down, the pages appear black

Hi, I am trying to create the assignment function: "substring<-" <- function(text, first, last=100000, sub) { if(is.character(first)) { if(!missing(last)) stop('wrong # arguments') return(sedit(text, first, sub)) } lf <- length(first) if(length(text)==1 && lf > 1) { if(missing(last)) last <- nchar(text) last <- rep(last,

Hi all, I am fancying the idea of using my preferred shell (t)csh when invoking system(). How can I do that? (I didn't find anything about that in the FAQ). Thanks for help, -christian Dr.sc.math.Christian W. Hoffmann Mathematics and Statistical Computing Landscape Modeling and Web Applications Swiss Federal Research Institute WSL Zuercherstrasse 111 CH-8903 Birmensdorf, Switzerland phone:

Hi, I want to use "pairs" on a fairly big matrix, say more than 30 columns. R tries to fit this on one page, which results in tiny invidual rectangles, and dots are hard to see. Is there a way to transform the output of pairs in a way that subplots of 10*10, say, rectangles are generated. These subplots should then go to Sweave which is accepting one figure at a time presently.

Hi all, I have a problem with plotting: (R-1.2.2 on SunOS fluke 5.5.1 Generic_103640-27 sun4u sparc SUNW,Ultra-Enterprise solaris 2.) I define a function to switch between plotting on graphics device and generating a postscript file (which can also be automatically be printed) plt <- function(view,f,file="") { if (view>0) f else {

Thanks for the tip to p.dalgaard and Prof Brian Ripley: For reasons that I have forgotten, the value of an assignment function is always passed using an argument called "value", so try changing "sub" to "value". The last argument of any assignment function has to be named `value'. This is different from S, and is in the FAQ, section 3.3.3: "In R, the last

Hi, I am trying to install ggobi for R under Solaris SunOS 5.5.1 Generic_103640-27 sun4u sparc SUNW,Ultra-Enterprise following the instructions in INSTALL under e) R interface, although I have some modifications in mind: I am planning to keep the sources (untarred and unzipped) in /home/woodstock/hoffmann/R/Sources/Rggobi and I have set setenv GGOBI_ROOT $HOME/R/Sources/Rggobi (with $HOME =

Hi all, Help! My help system is not functioning: >help.start() Making links in ~/.R ... If netscape is already running, it is *not* restarted, and you must switch to its window. Otherwise, be patient.. > C locale package not found. Using C locale by default. The netscape never comes up .. What could be wrong here? I have been experimenting with getting Rggobi running which may have

Hi, I am getting masked objects and I think the maintainers of the respective packages (lattice, boot, survival, aml) should agree on how to eliminate th conflicts. Here my start up protocol with the relevant information: -------------- R : Copyright 2005, The R Foundation for Statistical Computing Version 2.2.0 (2005-10-06 r35749) [clip] Type 'q()' to quit R. [clip] [Previously

Hi I have been unable to find information on the following problem: All the text in my *R* window is shown in light red, starting after the echo "This is mgcv 1.5-6 . For overview type ", the following `help("mgcv-package")' is already in the light red. This color annoys me because I need more contrast, and it is even worse for use with a beamer. In the options menu

Hi Sorry, I forgot to state the version: GNU Emacs 22.3.1 I have been unable to find information on the following problem: All the text in my *R* window is shown in light red, starting after the echo "This is mgcv 1.5-6 . For overview type ", the following `help("mgcv-package")' is already in the light red. This color annoys me because I need more contrast, and it is

Hi all, This seems to be a temporary problem of our local Solaris machine. Sorry for bugging you. --christian >Help! My help system is not functioning: >>help.start() >Making links in ~/.R ... >If netscape is already running, it is *not* restarted, >and you must switch to its window. >Otherwise, be patient.. >> C locale package not found. Using C locale by default.

Hi all, Using R Version 1.5.0 (2002-04-29) under Solaris I am reinstalling own packages useing the script #!/bin/csh cd ~/R/Sources R CMD check --no-latex --no-perms --no-install $1 R CMD INSTALL --library=lib $1 R CMD Rdindex $1 > INDEX R CMD build --force $1 Everything echoes with OK, *but* the files CONTENTS and TITLE will not be generated. What has changed in the process of