Re: [R] OC curve in Quality Control
WilDscOp [EMAIL PROTECTED] wrote: Can anyone please help me about any of the following questions: 1. How can i find factorial of any number in R? I tried prod(170:1) # to find factorial of 170 or 170! Is it the only procedure - or R has any better process / operational character to calculate factorial? Also, is it possible to calculate factorial of 500? Or is there any statistical table available for this? gamma(x+1) # gives x! As for the factorial of 500, I would go for a log transformation: lgamma(501) [1] 2611.330 But if you really need to see all the digits of 500!, PARI http://pari.math.u-bordeaux.fr/ --or rather its shell, may be called from within R. It returns all digits of that large number in a fraction of a second on my old laptop: system.time(system(echo '500!' | gp)) [...] %1 = 12201368259911100687012387854230469262535743428031928421 9241358838584537315388199760549644750220328186301361647714820 [about a thousand other digits follow] Good bye! [1] 0.00 0.00 0.05 0.03 0.02 For a more readable output: system(echo '500!*1.0' | gp) [...] %1 = 1.220136825991110068701238785 E1134 2. Is there any direct procedure / package in R to find permutation / combination? To permute the elements of a vector, you may use: perm - function(v)sample(v,size=length(v),replace=FALSE) v - 1:8 v [1] 1 2 3 4 5 6 7 8 set.seed(101) perm(v) [1] 3 1 5 4 7 6 2 8 Re combinations, does the function choose() do what you are looking for? 3. Can i find Probability Density of Hypergeometric Disribution in R, given all values of the parameter? I can not find any table of Hypergeometric Disribution (if such tables are available on the internet, please let me know). Does dhyper() do what you are looking for? -- Philippe Glaziou Epidemiologist __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Sweave question
Rafael A. Irizarry wrote: Using Sweave in the tools library (R version 1.8.0: sorry i havent upgraded), it seems i cant use if statements in R chunks that make graphs. i have this: fig=TRUE,echo=F= par(mfrow=c(1,1)) if(exists(x)) plot(x,x) else{ plot(1,1,type=n) text(1,1,data not available.\n) } @ and I get this error: Error: chunk 6 Error in parse(file, n, text, prompt) : parse error any help is appreciated. Whenever you get an error message, copy and paste it into an R session (interactive) and see what happens. At least then you'll know where the problem is, and whether it's a Sweave problem or a problem with your R syntax (in this case, it's the second option. sorry). This problem is almost, but not quite, a FAQ. The problem is that the if(...) is syntactically complete after the first plot() command. The else { appears out of nowhere, as far as the R interpreter is concerned, and this makes no sense. The fixes: 1) Always do this: if(something) { do something } else { do something else } Note exactly (!) where the curly braces are, relative to the else. 2) wrap the whole thing in curly braces. { if(something) do somthing else { do something else } } Cheers Jason -- Indigo Industrial Controls Ltd. http://www.indigoindustrial.co.nz 64-21-343-545 [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Plot a sphere
Spencer Graves wrote: A hemisphere is relatively easy; try the following: x - seq(-1, 1, length=21) Z - outer(x, x, function(x, y)sqrt(1-x^2-y^2)) persp(x=x, y=x, z=Z) A contour plot is also relatively easy: image(x=x, y=x, z=Z) contour(x=x, y=x, z=Z, add=T) However, if you want an honest perspective plot of a sphere complete with the underside, etc., I know of nothing in R that could do that. S-Plus has perspp, which could be used. However, that seems to be one of the few features available in S-Plus that is not currently available in R. The R package rgl by Adler and Nenadic can plot spheres. It is available at http://wsopuppenkiste.wiso.uni-goettingen.de/~dadler/rgl/ --- and looks like it will shortly become a CRAN package. Uwe Ligges hope this helps. spencer graves Derick Schoonbee wrote: Hi, I'm new to R (and math ;) Would somebody please be so kind as to direct me in plotting a 3D sphere? I tried something in the lines of: y - x - seq(-pi, pi, length=pi*10) f - function(x,y) { z - sqrt(pi - x^2 - y^2) #z[is.na(z)] - 0 z } z - outer(x, y, f) persp(x, y, z, theta = 120, phi = 30) I've also tried: make.surface.grid(...) .. persp( as.surface( grid, z) ) ... with the same result: 'Incomplete' demi sphere and others.. Any suggestions/solutions would be appreaciated. Regards, Derick PS:Merry X-mas ;) __ [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 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Plot a sphere
Thanks for the reply. The results are exactly the same as what I'm getting. Now I'm thinking in the lines of: z1 - outer(x, y, f) z2 - -outer(x, y, f) So, I want to attach the two hemispheres. But then I need to figure out how append vectors z1 z2 and then 'feed' this to persp.. hmm, I'll look into it. Regards, D. From: Spencer Graves [EMAIL PROTECTED] To: Derick Schoonbee [EMAIL PROTECTED] CC: [EMAIL PROTECTED] Subject: Re: [R] Plot a sphere Date: Thu, 25 Dec 2003 17:38:01 -0800 A hemisphere is relatively easy; try the following: x - seq(-1, 1, length=21) Z - outer(x, x, function(x, y)sqrt(1-x^2-y^2)) persp(x=x, y=x, z=Z) A contour plot is also relatively easy: image(x=x, y=x, z=Z) contour(x=x, y=x, z=Z, add=T) However, if you want an honest perspective plot of a sphere complete with the underside, etc., I know of nothing in R that could do that. S-Plus has perspp, which could be used. However, that seems to be one of the few features available in S-Plus that is not currently available in R. hope this helps. spencer graves Derick Schoonbee wrote: Hi, I'm new to R (and math ;) Would somebody please be so kind as to direct me in plotting a 3D sphere? I tried something in the lines of: y - x - seq(-pi, pi, length=pi*10) f - function(x,y) { z - sqrt(pi - x^2 - y^2) #z[is.na(z)] - 0 z } z - outer(x, y, f) persp(x, y, z, theta = 120, phi = 30) I've also tried: make.surface.grid(...) .. persp( as.surface( grid, z) ) ... with the same result: 'Incomplete' demi sphere and others.. Any suggestions/solutions would be appreaciated. Regards, Derick PS:Merry X-mas ;) __ [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: [R] Problems converting output from Sweave to PDf
Thon de Boer wrote: I am having trouble converting the output from Sweave into a valid PDF file. I have created a simple .Rnw file which will become a full vignette at some point, but during the intermediate testing, I got errors from texi2dvi. This is what I have done. 0) Using a Windows Xp system 1) Created a file called GeneSpring.Rnw 2) Convert this to Tex using Sweave(GeneSpring.Rnw) from within R 3) Since I don't have LATEX for windows, I used texi2dvi from the CYGWIN package 4) When I run the command in Cygwin 'texi2dvi -c -p GeneSpring.tex' I get the error shown below. 5) When I edit the GeneSpring.tex file and comment out the line \usepackage{C:/PROGRA~1/SILICO~1/GENESP~1/data/rw1081/share/texmf/Sweave} Well, cygwin misinterprets the path. Specify it in a way cygwin understands, e.g.: /cygdrive/c/. or even better, install a LaTeX environment for Windows. Uwe Ligges and try to convert to PDF again, it will successfully convert the TEX file to PDF. It seems that the Sweave library causes some problems in the parsing of the TEX file. Is there something I forgot to do? Thanks, Thon - OUTPUT FROM texi2dvi - /usr/local/bin:/usr/bin:/bin:/usr/X11R6/bin:/cygdrive/c/Perl/bin/:/cygdrive/c/WINDOWS/system32:/cygdrive/c/WINDOWS:/cygdrive/c/WINDOWS/System32/Wbem:/cygdrive/c/Program Files/ATI Technologies/ATI Control Panel:/cygdrive/c/Program Files/Common Files/Adaptec Shared/System:/cygdrive/c/Program Files/TEC100/BIN:/cygdrive/c/Program Files/bin:/cygdrive/c/Program Files/SiliconGenetics/GeneSpring/data/rw1081/bin This is pdfTeXk, Version 3.14159-1.10b (Web2C 7.4.5) %-line parsing enabled. (/tmp/GeneSpring.tex{/usr/share/texmf/pdftex/config/pdftex.cfg} LaTeX2e 2001/06/01 Babel v3.7h and hyphenation patterns for american, french, german, ngerman, n ohyphenation, loaded. (/usr/share/texmf/tex/latex/base/article.cls Document Class: article 2001/04/21 v1.4e Standard LaTeX document class (/usr/share/texmf/tex/latex/base/size10.clo)) ! Missing \endcsname inserted. to be read again \protect l.9 \begin {document} ? ! LaTeX Error: Missing \begin{document}. See the LaTeX manual or LaTeX Companion for explanation. Type H return for immediate help. ... l.9 \begin {document} ? -- TEX FILE -- % \VignetteIndexEntry{GeneSpring contents} % \VignetteDepends{GeneSpring} % \VignetteKeywords{Interface} % \VignettePackage{GeneSpring} \documentclass{article} \usepackage{C:/PROGRA~1/SILICO~1/GENESP~1/data/rw1081/share/texmf/Sweave} \begin{document} \author{Thon de Boer} \title{Using R with GeneSpring} \maketitle \copyright{2003 Silicon Genetics} \section{Introduction} This package contains a number of functions to facilitate the integration of R code into the Gene Expression analysis program GeneSpring, by Silicon Genetics. Available functions include: \begin{itemize} \item \texttt{GSload.int()} - Read GeneSpring experiment interpretation from file and return a GeneSpring gene expression object (GSint) \item \texttt{GSload.intBC()} - Read GeneSpring experiment interpretation from file and return a BioConductor gene expressioon object (exprSet) \item \texttt{GSload.exp()} - Read GeneSpring experiment from file and return a GeneSpring gene expression object (GSint) \item \texttt{GSload.expBC()} - Read GeneSpring experiment from file and return a BioConductor gene expressioon object (exprSet) \item \texttt{GSint()} - Create a GeneSpring gene expression object (GSint) \item \texttt{GSint2BC()} - Convert a GeneSpring gene expression object (GSint) to a BioConductor gene expression object (exprSet) \item \texttt{BC2GSint()} - Convert a BioConductor gene expression object (exprSet) to a GeneSpring gene expression object (GSint) \item \texttt{GSload.genelist()} - Read a GeneSpring GeneSpring gene list from file \item \texttt{GSsave.genelist()} - Save a GeneSpring GeneSpring gene list to file \item \texttt{GSsave.exp()} - Save a GeneSpring gene expression object (GSint) to file \end{itemize} For more information on using thi spackage with GeneSpring, go to the Silicon Genetics website. \end{document} __ [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
[R] Plot a sphere
To plot the sphere in R I find the best scatterplot3d package as follows: library(scatterplot3d) a=seq(-pi,pi, length=100) x=c(rep(1, 100) %*% t(cos(a))) y=c(cos(a) %*% t(sin(a))) z=c(sin(a) %*% t(sin(a))) scatterplot3d(x, y, z, type=l) Best regards Branimir K. Hackenberger __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] loop and read.table
Hi, I would like to open several tables with a loop, using something like : - $ ls 1.txt 2.txt 3.txt 4.txt $ R for (i in 1:4) tabi-read.table(i.txt) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `i.txt' -- thanks for any help Tristan Lefebure __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] loop and read.table
lefebure tristan wrote: Hi, I would like to open several tables with a loop, using something like : - $ ls 1.txt 2.txt 3.txt 4.txt $ R for (i in 1:4) tabi-read.table(i.txt) Since i is within a character string, it cannot be used as a variable in your case. You may paste() is together: paste(i, .txt, sep=), or even better, read the directories contents with list.files() and proceed over the result as in: tab - lapply(list.files(pattern=^?[[:digit:]]\.txt), read.table) Uwe Ligges Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `i.txt' -- thanks for any help Tristan Lefebure __ [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: [R] loop and read.table
On Fri, 26 Dec 2003, lefebure tristan wrote: I would like to open several tables with a loop, using something like : - $ ls 1.txt 2.txt 3.txt 4.txt $ R for (i in 1:4) tabi-read.table(i.txt) Use read.table(paste(i, txt, sep=.)) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `i.txt' -- -- 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] Multiple dependent variables
Dear friends, I'm stuck with the following problem: I would like to do a multiple regression with muliple dependent variables, and i don't know how to write my formula in the model. Someone in previous messages offered the cbind method, but the result is just as many regression as the number of dependent variables, it is just a time saving method; i'm looking for a method closer to a MANOVA, but without factors. well, a standard multiple rgression with multiple dependent variable, i.e. looking for variables that are significant in all the dependent variables. If one of you have an idea, I welcome it. and happy new year to all simon [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Plot a sphere
Derick Schoonbee [EMAIL PROTECTED] wrote: Would somebody please be so kind as to direct me in plotting a 3D sphere? Here's one way. I generate an empty 3D plot with persp, then fill it with polygons transformed with trans3d (as found in the help for persp). I didn't do hidden surface removal (you didn't mention whether you wanted it), but if you do, just reorder the polygons from back to front and paint them a solid color (e.g. col=red), so hidden ones get painted over. pmat - persp(0:1, 0:1, matrix(,2,2), xlim=c(-1,1), ylim=c(-1,1), zlim=c(-1,1), theta=25, phi=30, expand=.9, xlab=X, ylab=Y, zlab=Z) trans3d - function(x,y,z, pmat) { # From the help for persp tr - cbind(x,y,z,1) %*% pmat list(x = tr[,1]/tr[,4], y= tr[,2]/tr[,4]) } theta - seq(0, 2*pi, length=51) phi - seq(0, pi, length=26) x - cos(theta) %o% sin(phi) y - sin(theta) %o% sin(phi) z - rep(1, length(theta)) %o% cos(phi) for (j in seq(phi)[-1]) for (i in seq(theta)[-1]) { idx - rbind(c(i-1,j-1), c(i,j-1), c(i,j), c(i-1,j)) polygon(trans3d(x[idx], y[idx], z[idx], pmat)) } -- -- David Brahm ([EMAIL PROTECTED]) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] loop and read.table
And if your intention is to create 4 data frames with names tab1, tab2, tab3, tab4 then combine Prof. Riley's advice with assign like this: for(i in 1:4) assign( paste(tab, i, sep=), read.table(paste(i, txt, sep=.)) ) --- Date: Fri, 26 Dec 2003 14:07:45 + (GMT) From: Prof Brian Ripley [EMAIL PROTECTED] To: lefebure tristan [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: Re: [R] loop and read.table On Fri, 26 Dec 2003, lefebure tristan wrote: I would like to open several tables with a loop, using something like : - $ ls 1.txt 2.txt 3.txt 4.txt $ R for (i in 1:4) tabi-read.table(i.txt) Use read.table(paste(i, txt, sep=.)) Error in file(file, r) : unable to open connection In addition: Warning message: cannot open file `i.txt' -- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] re| Dr Ward on List protocol
Andrew C. Ward [EMAIL PROTECTED] : With respect to 'tone' and 'friendliness', perhaps all that is meant or needed is that people be polite and respectful. I shake my head as often at rude answers Oh, by gosh, by golly. I don't think an occasional dose of 'real life', via a jab from the Professor, will cause any lasting harm to the cosseted emolumated students and academics on the List. On a Wall St trading desk, for example, every day one is kicked in the head more brutally by clients, superiors, counterparts, the markets etc, than ever one would be by the Professor. Plus, the Professor's jabs are good Schadenfreudic fun for the rest of us. . Regards, Steve Wisdom Westport CT US ps/ In the interest of full disclosure, I own four of the Professor's books __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] Multiple dependent variables
Simon CHAMAILLE [EMAIL PROTECTED] writes: Dear friends, I'm stuck with the following problem: I would like to do a multiple regression with muliple dependent variables, and i don't know how to write my formula in the model. Someone in previous messages offered the cbind method, but the result is just as many regression as the number of dependent variables, it is just a time saving method; i'm looking for a method closer to a MANOVA, but without factors. well, a standard multiple rgression with multiple dependent variable, i.e. looking for variables that are significant in all the dependent variables. If one of you have an idea, I welcome it. and happy new year to all simon Some of this stuff appears to be not quite implemented, but have a look at summary(manova(cbind(y1,y2)~x1+x2)) Df Pillai approx F num Df den Df Pr(F) x1 1 0.079014 0.257378 2 6 0.7812 x2 1 0.006782 0.020485 2 6 0.9798 Residuals 7 summary(manova(cbind(y1,y2)~x2+x1)) Df Pillai approx F num Df den Df Pr(F) x2 1 0.006930 0.020936 2 6 0.9794 x1 1 0.078886 0.256927 2 6 0.7815 The annoying thing is that drop1 and friends are not there, so you have to arrange terms so that the sequential tests make sense. However, all the bits are clearly there. Another annoying thing is that manova() is not analogous to anova(). -- 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] re| Dr Ward on List protocol
Steve Wisdom wrote: I don't think an occasional dose of 'real life', via a jab from the Professor, will cause any lasting harm to the cosseted emolumated students and academics on the List. On a Wall St trading desk, for example, every day one is kicked in the head more brutally by clients, superiors, counterparts, the markets etc, than ever one would be by the Professor. Plus, the Professor's jabs are good Schadenfreudic fun for the rest of us. U, there's more than ***one*** rude professor who contributes to the list. Although I play in a very minor league academically, I like to think that my capacity for rudeness is right up there with the best of them! :-) cheers, Rolf Turner [EMAIL PROTECTED] P. S. My New Oxford Dictionary lists the word ``emolument'' as a noun, but does not list a verb ``emolumate''. (???) R. T. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Dr Ward on List protocol
Shouldn't posters who do not do obvious research before asking their questions have this forcefully pointed out to them? I think they should. It may put people off posting, but it may also make them do the work. As a longtime lurker who pays a lot of money for good statistical advice, I have an idea of what it would cost in the real world to have questions answered by the people on this list, though the money is not the only measure, I know. So it's not a resource that should be treated too lightly, and I think the onus should be on the questioners rather than the respondents to take this on board. Seems to me the list has worked pretty well so far, and I frankly don't see the need for a change. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] Re: OC curve in Quality Control
Dear All, First of all, thanks to Philippe Glaziou and Andrew C. Ward for their nice and very helpful e-mail. Well.. - --- For Number 1 problem: about http://pari.math.u-bordeaux.fr/, sorry i didnot understand what to download or what to call! From http://pari.math.u-bordeaux.fr/download.html can you please tell me what should i download and how to use(i could not find any doc)? An exact link for download may be helpful:) - --- For Number 2 problem: perm function is a very nice one. And yes choose() works fine for me. Also gregmisc package http://cran.r-project.org/bin/windows/contrib/1.9/gregmisc_0.8.7.zip is good. - --- For Number 3 problem: i had some problem using dhyper() because i cannot match the parameters there. I used N-5000; n-50; D-seq(0,500,1); d-0 # (For d = 1 i just recalculated it with d-1) hypg - (choose(D, d)* choose((N-D), (n-d)) / choose(N,n)) instead and it surves my purpose. Anyway, is there any better procedure? I could not find Hypergeometric() in the base. - --- For Number 5 problem: I did as follows: # For type A OC curve N-5000; n-50; D-seq(1,500,1); d1-0; d2-1; p-(seq(1,500,1)/5000); Pa-((choose(D, d1)* choose((N-D), (n-d1)) / choose(N,n)) + (choose(D, d2)* choose((N-D), (n-d2)) / choose(N,n))); plot(p,Pa,pch=o, main=Type A \nOperating Characteristic Curve, sub=Using Hypergeometric Distribution, xlab=Proportion defective, ylab=Probability of Acceptance, xlim=c(0.00,0.10), ylim=c(0.00,1.00), type=p, axes =T, col=1) # here, p = D/N # For type B OC curve N-5000; n-50; D-seq(0,500,1); d1-0; d2-1; p-(seq(0,500,1)/5000); Pa1- dpois(0, (n*p), log = FALSE); Pa2- dpois(1, (n*p), log = FALSE); Pa-Pa1+Pa2; plot(p,Pa,pch=o, main=Type B \nOperating Characteristic Curve, sub=Using Poisson as an approximation of Binomial Distribution, xlab=Proportion defective, ylab=Probability of Acceptance, xlim=c(0.00,0.10), ylim=c(0.00,1.00), type=p, axes =T, col=1) And these works fine. Thank you for your time. ___ Mohammad Ehsanul Karim [EMAIL PROTECTED] Institute of Statistical Research and Training University of Dhaka, Dhaka- 1000, Bangladesh __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
RE: [R] draft of posting guide
Rather than a separate beginners' mailing list or a posting guide, perhaps what we need is a separate mailing list for discussing posting style? -thomas __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] different results by re-ordering vector: bug?
Dear R users, I obtain the following behavior which I cannot understand. Given a file mydata with the following numbers: 0.171409662475182 0.15817339510258108 0.32230311052283256 0.14890800794176043 0.17074784910655194 0.16611515552614162 0.41 0.16611515552614162 0.41760423560555926 0.11978821972203839 I read the data and perform some calculations: a - 1-read.table(mydata)$V1 m - outer(a, a, /) diag(m) - NA mean.row - apply(m, 1, mean, na.rm=TRUE) which yield the same value for indices 6 and 8 of mean.row, as would be expected because values 6 and 8 of the original vector are the same: mean.row[6]==mean.row[8] [1] TRUE However, if I reorder the values as follows: a - 1-read.table(mydata)$V1[c(10,2,8,9,7,3,1,4,5,6)] and repeat the calculations: m - outer(a, a, /) diag(m) - NA mean.row - apply(m, 1, mean, na.rm=TRUE) mean.row[6]==mean.row[8] The values for indices 10 and 3 of mean.row, which correspond to 6 and 8 in the previous calculations, are not the same anymore: mean.row[10]==mean.row[3] [1] FALSE I understand that limited precision causes incorrect results but I wouldn't expect ordering operations to do the same. I couldn't find any information in the site about this. Maybe it's a bug with my version: R.version _ platform i686-pc-linux-gnu arch i686 os linux-gnu system i686, linux-gnu status major1 minor7.0 year 2003 month04 day 16 language R Thanks and best regards, Carlos __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] different results by re-ordering vector: bug?
I just got the same answer both times with R 1.8.1 under Windows 2000 on an IBM Thinkpad T30. However, if abs(diff(mean.row[c(3,10)])) is not much bigger than .Machine$double.esp*sum(abs(mean.row[c(3,10)])), then I would not call that a bug. Rather, it should be considered a warning not to expect exact equality in comparing floating point numbers Just now, I checked (pi+x-x) == 4*atan(1) and got TRUE for both in R 1.8.1 and S-Plus 2000 with x = 1 but FALSE with x = 100. hope this helps. spencer graves Carlos Soares wrote: Dear R users, I obtain the following behavior which I cannot understand. Given a file mydata with the following numbers: 0.171409662475182 0.15817339510258108 0.32230311052283256 0.14890800794176043 0.17074784910655194 0.16611515552614162 0.41 0.16611515552614162 0.41760423560555926 0.11978821972203839 I read the data and perform some calculations: a - 1-read.table(mydata)$V1 m - outer(a, a, /) diag(m) - NA mean.row - apply(m, 1, mean, na.rm=TRUE) which yield the same value for indices 6 and 8 of mean.row, as would be expected because values 6 and 8 of the original vector are the same: mean.row[6]==mean.row[8] [1] TRUE However, if I reorder the values as follows: a - 1-read.table(mydata)$V1[c(10,2,8,9,7,3,1,4,5,6)] and repeat the calculations: m - outer(a, a, /) diag(m) - NA mean.row - apply(m, 1, mean, na.rm=TRUE) mean.row[6]==mean.row[8] The values for indices 10 and 3 of mean.row, which correspond to 6 and 8 in the previous calculations, are not the same anymore: mean.row[10]==mean.row[3] [1] FALSE I understand that limited precision causes incorrect results but I wouldn't expect ordering operations to do the same. I couldn't find any information in the site about this. Maybe it's a bug with my version: R.version _ platform i686-pc-linux-gnu arch i686os linux-gnu system i686, linux-gnu status major1 minor 7.0 year 2003month04 day 16 language R Thanks and best regards, Carlos __ [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: [R] Re: OC curve in Quality Control
WilDscOp [EMAIL PROTECTED] wrote: For Number 1 problem: about http://pari.math.u-bordeaux.fr/, sorry i didnot understand what to download or what to call! From http://pari.math.u-bordeaux.fr/download.html can you please tell me what should i download and how to use(i could not find any doc)? An exact link for download may be helpful:) The download page gives links to the sources of PARI. There is also a self-installing MS-Windows executable there. As for the doc, there is a menu item which reads Documentation. Click on it, and there you have a user's guide, tutorial, installation guide, ref card, manual pages. By the way, there are other packages out there to handle large entities (on a 32-bit machine) like 500! (Octave, Matlab, Maple, Mathematica, Yacas...). I suggested that one because I do not know the others. - --- For Number 3 problem: i had some problem using dhyper() because i cannot match the parameters there. I used N-5000; n-50; D-seq(0,500,1); d-0 # (For d = 1 i just recalculated it with d-1) hypg - (choose(D, d)* choose((N-D), (n-d)) / choose(N,n)) instead and it surves my purpose. Anyway, is there any better procedure? I could not find Hypergeometric() in the base. You needed phyper for that specific problem, not dhyper (the help page for dhyper explains what dhyper, phyper, qhyper and rhyper do). hypg - phyper(d,D,N-D,n) does the job in a slightly more efficient manner. I would suggest that you check out the following: help.search(hypergeometric) ?Hypergeometric -- Philippe Glaziou Epidemiologist __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help