[R] Indexing ANOVA table
Hi all, I'd like to extract a value from an ANOVA table, but experience the following problem: ### This works: s.pseudo - summary(aov(m ~ block + mix*graz,data=split1)) s.pseudo Df Sum Sq Mean Sq F value Pr(F) block2 1114.66 557.33 4.4296 0.04192 * mix 16.146.14 0.0488 0.82956 graz 21.450.72 0.0057 0.99427 mix:graz 23.821.91 0.0152 0.98495 Residuals 10 1258.19 125.82 --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 s.pseudo[[1]][mix ,Pr(F)] [1] 0.8295556 ### But this doesn't -- why? s.split - summary(aov(m ~ block + mix*graz + Error(Plot),data=split1)) s.split Error: Plot Df Sum Sq Mean Sq F value Pr(F) block 2 1114.66 557.33 0.8994 0.5265 mix16.146.14 0.0099 0.9298 Residuals 2 1239.37 619.68 Error: Within Df Sum Sq Mean Sq F value Pr(F) graz 2 1.4464 0.7232 0.3073 0.7437 mix:graz 2 3.8206 1.9103 0.8117 0.4776 Residuals 8 18.8278 2.3535 s.split[[Error: Plot]] ## extracting first list element works Df Sum Sq Mean Sq F value Pr(F) block 2 1114.66 557.33 0.8994 0.5265 mix16.146.14 0.0099 0.9298 Residuals 2 1239.37 619.68 s.split[[Error: Plot]][mix ,Pr(F)] ### == FAILS Error in s.split[[Error: Plot]][mix , Pr(F)] : incorrect number of dimensions So where is the difference between the two? Thanks for any hint Pascal - This mail sent through IMP: http://horde.org/imp/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] significance in difference of proportions: What problema
On 28-Nov-03 Torsten Hothorn wrote: yes, thats my understanding too. The enumerative techniques as you call it condition on the data actually observed and determine the null distribution of the associated test statistic from the data. In contrast, unconditional procedures require some assumptions to the underlying data generating process from which the null distribution is derived. The appropriate choice depends of the kind of experiment under test: In a randomized trial we would like to see all possible outcomes of the trial caused by rerandomization and the enumerative techniques are natural here. When we draw many samples from predefined populations, men and women, say, rerandomization of gender is of course not that easy and we may assume something about the data generating process :-) Nice example, but it depends on how you look at it! Suppose you have samples of n1 Men and n2 Women and record, for instance, whether or not each is suffering from a cold (r1 and r2 respectively). Do M W differ in their risk of catching cold? NH: No difference; implies that the R = (r1+r2) colds have selected a random subset of the N=(n1+n2) individuals as victims; implies that the n1 Men out of N are a random subset of the R+(N-R) Colds/NonColds. So you then have a hypergeometric distribution and are back with an exact test. But are we assuming somthing about the data generating process here? (Of course, in the background lurks the Ogre of Exchangeability, in that the probability of catching cold may vary from person to person, whether Man or Woman, but nothing in the information plus NH suggests any reason to distinguish any arrangement of the N people from any other; equivalent to a re-randomisation of gender ... ??). Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 167 1972 Date: 29-Nov-03 Time: 10:09:18 -- XFMail -- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Indexing ANOVA table
[EMAIL PROTECTED] writes: Hi all, I'd like to extract a value from an ANOVA table, but experience the following problem: ### This works: s.pseudo - summary(aov(m ~ block + mix*graz,data=split1)) s.pseudo Df Sum Sq Mean Sq F value Pr(F) block2 1114.66 557.33 4.4296 0.04192 * mix 16.146.14 0.0488 0.82956 graz 21.450.72 0.0057 0.99427 mix:graz 23.821.91 0.0152 0.98495 Residuals 10 1258.19 125.82 --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 s.pseudo[[1]][mix ,Pr(F)] [1] 0.8295556 ### But this doesn't -- why? s.split - summary(aov(m ~ block + mix*graz + Error(Plot),data=split1)) s.split Error: Plot Df Sum Sq Mean Sq F value Pr(F) block 2 1114.66 557.33 0.8994 0.5265 mix16.146.14 0.0099 0.9298 Residuals 2 1239.37 619.68 Error: Within Df Sum Sq Mean Sq F value Pr(F) graz 2 1.4464 0.7232 0.3073 0.7437 mix:graz 2 3.8206 1.9103 0.8117 0.4776 Residuals 8 18.8278 2.3535 s.split[[Error: Plot]] ## extracting first list element works Df Sum Sq Mean Sq F value Pr(F) block 2 1114.66 557.33 0.8994 0.5265 mix16.146.14 0.0099 0.9298 Residuals 2 1239.37 619.68 s.split[[Error: Plot]][mix ,Pr(F)] ### == FAILS Error in s.split[[Error: Plot]][mix , Pr(F)] : incorrect number of dimensions So where is the difference between the two? Thanks for any hint There's an extra list level: example(aov) ... str(summary(npk.aovE)[[Error: Within]]) List of 1 $ :Classes anova and `data.frame':7 obs. of 5 variables: ..$ Df : num [1:7] 1 1 1 1 1 1 12 ..$ Sum Sq : num [1:7] 189.3 8.4 95.2 21.3 33.1 ... ..$ Mean Sq: num [1:7] 189.3 8.4 95.2 21.3 33.1 ... ..$ F value: num [1:7] 12.259 0.544 6.166 1.378 2.146 ... ..$ Pr(F) : num [1:7] 0.00437 0.47490 0.02880 0.26317 0.16865 ... - attr(*, class)= chr [1:2] summary.aov listof summary(npk.aovE)[[Error: Within]][[1]][N:P,Pr(F)] [1] 0.2631653 As to *why* there's this extra level, you have to ask the author... -- 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] Indexing ANOVA table
On Sat, 29 Nov 2003 [EMAIL PROTECTED] wrote: Hi all, I'd like to extract a value from an ANOVA table, but experience the following problem: ### This works: s.pseudo - summary(aov(m ~ block + mix*graz,data=split1)) s.pseudo Df Sum Sq Mean Sq F value Pr(F) block2 1114.66 557.33 4.4296 0.04192 * mix 16.146.14 0.0488 0.82956 graz 21.450.72 0.0057 0.99427 mix:graz 23.821.91 0.0152 0.98495 Residuals 10 1258.19 125.82 --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 s.pseudo[[1]][mix ,Pr(F)] [1] 0.8295556 ### But this doesn't -- why? Why should it? s.split - summary(aov(m ~ block + mix*graz + Error(Plot),data=split1)) s.split Error: Plot Df Sum Sq Mean Sq F value Pr(F) block 2 1114.66 557.33 0.8994 0.5265 mix16.146.14 0.0099 0.9298 Residuals 2 1239.37 619.68 Error: Within Df Sum Sq Mean Sq F value Pr(F) graz 2 1.4464 0.7232 0.3073 0.7437 mix:graz 2 3.8206 1.9103 0.8117 0.4776 Residuals 8 18.8278 2.3535 s.split[[Error: Plot]] ## extracting first list element works Df Sum Sq Mean Sq F value Pr(F) block 2 1114.66 557.33 0.8994 0.5265 mix16.146.14 0.0099 0.9298 Residuals 2 1239.37 619.68 s.split[[Error: Plot]][mix ,Pr(F)] ### == FAILS Error in s.split[[Error: Plot]][mix , Pr(F)] : incorrect number of dimensions You should have written s.split[[Error: Plot]][[1]][mix ,Pr(F)] So where is the difference between the two? One is an aov object, the other an aovlist object. Take a closer look at print.summary.aov, for example. More generally, learn how to look at R objects instead of assuming you know what you are doing: unclass(s.split[[Error: Plot]]) would have been informative. -- 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] performance gap between R 1.7.1 and 1.8.0
AndyL == Liaw, Andy [EMAIL PROTECTED] on Fri, 28 Nov 2003 22:19:27 -0500 writes: AndyL Dear R-help, A colleague of mine was running some AndyL code on two of our boxes, and noticed a rather large AndyL difference in running time. We've so far isolated AndyL the problem to the difference between R 1.7.1 and AndyL 1.8.0, but not more than that. The exact same code AndyL took 933.5 seconds in 1.7.1, and 3594.4 seconds in AndyL 1.8.1, on the same box. I understand your concern. I do believe that it must be some peculiarity of your (or the non-standard packages you used code). Some wild guessing: Looking at your profiling table, and seeing read.dcf() makes me think about library() -- which does call read.dcf() only when called on a package that's not yet attached --- and the bioconductor extra package management -- which for me makes only the first time (!) attachment of bioconductor packages relatively slow. I still have no idea how this could influence bootstrapping, unless you detach() and attach() packages many times. In any case, it might depend on the exact list (and order!?) of packages present in search() when calling your code. A very first step of diagnosis might be to activate trace(read.dcf) trace(library) options(verbose = TRUE) A step further might be to patch read.dcf such that it prints info if(getOption(verbose)) { print what I am trying to read } Martin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] MASS fitdistr()
At 15:49 27-11-2003, you wrote: Dear R experts, I am trying to use the R MASS library fitdistr() to fit the following list: k21stsList-c(0.76697,0.57642,0.75938,0.82616,0.93706,0.77377,0.58923,0.37157,0.60796,1.00070,0.97529,0.62858,0.63504,0.68697,0.61714,0.75227,1.16390,0.66702,0.83578) as follows, library(MASS) fitdistr(k21stsList, normal) But, I get Error in fitdistr(k21stsList, normal) : 'start' must be a named list Hi You donĀ“t put the start list, something like this fitdistr(k21stsList, normal,list(mean = 0.5 sd = 0.1)) mean sd 0.74584591 0.17908744 (0.04108548) (0.02904255) Bernardo Rangel Tura, MD, MSc National Institute of Cardiology Laranjeiras Rio de Janeiro Brazil __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] problems with xlab
Dear R-list members, I'm using the 'effects' package to plot a graph with fixed values. For some reason, I cannot change the label for the 'x' axis using 'xlab='. The usual commands work for the y axis and for the main title. My last attempt used the following syntax: model.effects-(all.effects(model.1)) plot (model.effects, main=Figure 1,xlab=Year,ylab=Change) The resulting graph has the default label for the x axis instead of 'Year'. The main title and the label for 'y' are as expected. Anyone knows what may be happening? All the Best, Celso Celso F. Rocha de Barros DPhil Candidate in Sociology Lady Margaret Hall, Oxford University __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] compile problem with gzfile
Try compiling after configure --without-zlib (available in a less obselete system than 1.7.1). This skips your system's zlib (which is apparently fragile) and compiles that in the R sources. On Sat, 29 Nov 2003, Thomas W Blackwell wrote: William - I don't have an answer for you, just some stream of consciousness rambling about how I would go about diagnosing this, if it were my problem. In my installation directory (of R-1.7.1 on linux) there is a file config.log which contains all of the configure script's queries and replies. Yes, there is a difference between configuring and compiling, but that file might be one place to start looking for where the problem begins. For me, 'grep gzfile config.log' returns no matching lines. Nor should it: it is lib/zlib that is being invoked, and gzfile() is an R function. However, 'grep methods config.log' returns three lines which say config.status:884: creating src/library/methods/DESCRIPTION config.status:884: creating src/library/methods/Makefile config.status:884: creating src/library/methods/src/Makefile Seems odd that the 'methods' package would be the only one which fails to unpack. Not at all. That is the only package that is installed as a compressed saved image rather than as a text source file. Perhaps the problem is more general, and, likely, more elementary. Ah. When you downloaded the source files, did you set 'binary' in ftp or equivalent ? HTH - tom blackwell - u michigan medical school - ann arbor - On Fri, 28 Nov 2003, William D. McCoy wrote: I just finished compiling today's r-patched (R 1.8.1 patched) on my old Sparc 20 using gcc 3.3 along with Sun's f77. As it was compiling, I noticed an error regarding gzfile. The compile process left the methods directory and continued on. Obviously, there was a problem, but I don't have the original error message from the compiler. After installing R, I get the following error message when I start R: Error in open.connection(con, rb) : unable to open connection In addition: Warning message: cannot open compressed file `/usr/local/lib/R/library/methods/R/all.rda' A check for the file 'all.rda' showed that it does not exist. I do have gzip installed and I have libz.so.1.1.4 in /usr/local/lib. I realize this isn't much to go on, but does anyone have an idea what might be wrong? Has anyone had a similar error? -- William D. McCoy Geosciences University of Massachusetts Amherst, MA 01003 __ [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 -- 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] problems with xlab
Dear Celso, At 04:52 PM 11/29/2003 +, Celso Barros wrote: Dear R-list members, I'm using the 'effects' package to plot a graph with fixed values. For some reason, I cannot change the label for the 'x' axis using 'xlab='. The usual commands work for the y axis and for the main title. My last attempt used the following syntax: model.effects-(all.effects(model.1)) plot (model.effects, main=Figure 1,xlab=Year,ylab=Change) The resulting graph has the default label for the x axis instead of 'Year'. The main title and the label for 'y' are as expected. Anyone knows what may be happening? I'm afraid that you've found a bug in plot.effect(): I omitted specifying the xlab argument to xyplot() in one of the six cases that plot.effect() distinguishes. The fix is simple, and I'll shortly upload a new version of the package with it, but in the meantime, I'll send you (privately) the fixed version of plot.effect(). I apologize for the problem, John - 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
Re: [R] compile problem with gzfile
Thanks to Brian Ripley and Thomas Blackwell for responses concerning my problem. I am following Brian's suggestion and recompiling after configuring with --without-zlib. The compile takes a couple of hours on the old Sparc 20, but the process has passed the point where it failed before and has successfully completed compiliation of the methods package. So it looks like it will be alright. Prof Brian Ripley writes: Try compiling after configure --without-zlib (available in a less obselete system than 1.7.1). This skips your system's zlib (which is apparently fragile) and compiles that in the R sources. On Sat, 29 Nov 2003, Thomas W Blackwell wrote: William - I don't have an answer for you, just some stream of consciousness rambling about how I would go about diagnosing this, if it were my problem. In my installation directory (of R-1.7.1 on linux) there is a file config.log which contains all of the configure script's queries and replies. Yes, there is a difference between configuring and compiling, but that file might be one place to start looking for where the problem begins. For me, 'grep gzfile config.log' returns no matching lines. Nor should it: it is lib/zlib that is being invoked, and gzfile() is an R function. However, 'grep methods config.log' returns three lines which say config.status:884: creating src/library/methods/DESCRIPTION config.status:884: creating src/library/methods/Makefile config.status:884: creating src/library/methods/src/Makefile Seems odd that the 'methods' package would be the only one which fails to unpack. Not at all. That is the only package that is installed as a compressed saved image rather than as a text source file. Perhaps the problem is more general, and, likely, more elementary. Ah. When you downloaded the source files, did you set 'binary' in ftp or equivalent ? HTH - tom blackwell - u michigan medical school - ann arbor - On Fri, 28 Nov 2003, William D. McCoy wrote: I just finished compiling today's r-patched (R 1.8.1 patched) on my old Sparc 20 using gcc 3.3 along with Sun's f77. As it was compiling, I noticed an error regarding gzfile. The compile process left the methods directory and continued on. Obviously, there was a problem, but I don't have the original error message from the compiler. After installing R, I get the following error message when I start R: Error in open.connection(con, rb) : unable to open connection In addition: Warning message: cannot open compressed file `/usr/local/lib/R/library/methods/R/all.rda' A check for the file 'all.rda' showed that it does not exist. I do have gzip installed and I have libz.so.1.1.4 in /usr/local/lib. I realize this isn't much to go on, but does anyone have an idea what might be wrong? Has anyone had a similar error? -- William D. McCoy Geosciences University of Massachusetts Amherst, MA 01003 __ [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 -- 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 -- William D. McCoy Geosciences University of Massachusetts Amherst, MA 01003 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Classic Levene Test of variances
Hello, I am searching for the classic Levene test of variances in R but I don't find the function. Is there any command or do I have to programm the test by myself? Does anybody know? Help would be fine because R is very new to me. S.Zank [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Classic Levene Test of variances
Sebastian Zank [EMAIL PROTECTED] writes: Hello, I am searching for the classic Levene test of variances in R but I don't find the function. Is there any command or do I have to programm the test by myself? Does anybody know? Help would be fine because R is very new to me. There's one inside the Rcmdr package. http://cran.r-project.org/src/contrib/PACKAGES.html#Rcmdr -- 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] tcltk problem with Button-2
The following R code works as expected: require( tcltk ) tt - tktoplevel() # create a button labelled A that changes to B when pressed pressed - function() tkconfigure( tt.but, text=B ) tt.but - tkbutton( tt, text=A, command=pressed ) tkpack(tt.but) # if Control-Button-1 pressed change button label to C e - expression( tkconfigure( tt.but, text=C ), break ) tkbind(tt.but, Control-Button-1, e ) But if I simply replace Control-Button-1 with Button-2 in the tkbind command, i.e. the last line, then nothing happens when I press Button-2 (which I assume is the right mouse button). I also tried removing the break and using a function instead of an expression but it also does not respond, i.e. I tried replacing the last line with: tkbind(tt.but, Button-2, function() tkconfigure( tt.but, text=C ) ) I am using Windows 2000 and R 1.8.1 and am pasting the script into Rterm to make sure that the Rgui does not interfere with it. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help