similar to: inconsistency in pchisq (PR#7099)

The function 'pchisq' from the 'stats' library gives a wrong result if the argument equals exactly zero: # Upper tail of central 1-df chi^2 distribution > pchisq(1 , 1, ncp=0, lower.tail = F, log.p = FALSE) [1] 0.3173105 > pchisq(0.5 , 1, ncp=0, lower.tail = F, log.p = FALSE) [1] 0.4795001 > pchisq(0.01 , 1, ncp=0, lower.tail = F, log.p = FALSE) [1]

Full_Name: Alan Bain Version: 2.4.0 OS: XP Submission from: (NULL) (155.140.122.227) Code for pnchisq contains following if (tSml) { if (x> f+ theta+ 5*sqrt(2*(f+2*theta) ))){ return 1.; /* better than 0 --- but definately FIXME */ } } This needs to check which tail has been requested; it is only correct if the default lower_tail=1 has been requested; for upper tail should return 0

I have dowloaded the Source Code of R,and I want to know the process of chi-sqared test,but how can I found it? [[alternative HTML version deleted]]

Full_Name: Jerry W. Lewis Version: 2.6.1 OS: Windows XP Professional Submission from: (NULL) (24.147.191.250) pchisq(0,0,ncp=lambda) returns 0 instead of exp(-lambda/2) pchisq(x,0,ncp=lambda) returns NaN instead of exp(-lambda/2)*(1 + SUM_{r=0}^infty ((lambda/2)^r / r!) pchisq(x, df + 2r)) qchisq(.7,0,ncp=1) returns 1.712252 instead of 0.701297103 qchisq(exp(-1/2),0,ncp=1) returns 1.238938

Full_Name: Drouilhet R?my Version: 1.8.1 OS: Linux Submission from: (NULL) (195.221.43.136) qchisq(1,10) works well but qchisq(1,10,ncp=0) does not work whereas ncp=0 is the default value of the function qchisq(1,10). (of course, 10 will be replaced by any integer value). Let us notice that this bug occurs only when applying probability one. (qchisq(seq(0,.9,.1),10,ncp=0) works very well).

Hi, Thanks, everyone for all the help! So, here is my calling function in C (called test.c): #include<stdio.h> #include<stdlib.h> #include<Rmath.h> int main(void) { printf("%f \n",pchisq(2.,7., 1, 0)); printf("%f \n",pnchisq(2.,7.,0., 1, 0)); return EXIT_SUCCESS; } I compile using: gcc test.c -I/usr/lib/R/include

Dear all Just for fun, I have just downloaded the paper mentioned below and checked it with R-1.6.1. Everything is ok with exception of Table 2b, where I get always 1 instead of 0.5: > pbinom(1e15,2e15,0.5) [1] 1 Which value should be correct? Best regards Christian Stratowa ============================================== Christian Stratowa, PhD Boehringer Ingelheim Austria Dept NCE Lead

Dear colleague, I not sure this R code is correctly ? I would to show the number of Sample Size at Sample Size Axis that line draw from Power Axis (80%) from R code. How I show this and select the most appropriate of this power (.79955687 - 80983575). Thank for your help and answer. Best Regards, Nikom Thanomsieng, Email: nikom at kku.ac.th .... #Power analysis: Sample size for

for example, an expression such as chisq(df=1,ncp=0) ? thanks -- View this message in context: http://www.nabble.com/how-to-calculate-chisq-value-in-R-tp15338943p15338943.html Sent from the R help mailing list archive at Nabble.com.

Dear R experts, I'm trying to call 'pchisq' from within a C subroutine. The following error is returned: ** NON-convergence in pgamma()'s pd_lower_cf() f= nan. This error message is not printed the first time I call 'pchisq' from the C subroutine, but the second time or the next time I call 'pchisq' from within R. My session output is shown below:

(1) ...need to speed up a monte-carlo sampling...any suggestions about how I can get R to use all 8 cores of a mac pro would be most useful and very appreciated... (2) spent the last few hours trying to get pnmath to compile under os- x 10.5.4... using gcc version 4.2.1 (Apple Inc. build 5553) as downloaded from CRAN, xcode 3.0... ...xcode 3.1 installed over top of above after

Hello all, Here's my query: Running R 1.6.2 on FreeBSD 5.0, and on WinXP, and I find that the following hangs the process: dchisq(alpha=0.01, df=1, ncp=295) but it does work for ncp < 294.92. Is this general? Best wishes to all, Andrew Andrew Robinson Ph: 208 885 7115 Department of Forest Resources Fa: 208 885 6226 University of Idaho E : andrewr at uidaho.edu PO

I compiled the 1.9.1 src.rpm with the standard gnu tools and it works. I tried compiling the 1.9.1 src.rpm with the Intel 8 C and FORTRAN compilers and it bombs out during the testing phase: comparing 'd-p-q-r-tests.Rout' to './d-p-q-r-tests.Rout.save' ...267c267 < df = 0.5[1] "Mean relative difference: 5.001647e-10" --- > df = 0.5[1] TRUE

Hi Folks, I note that, while the "chisq" functions dchisq(x, df, ncp=0, log = FALSE) pchisq(q, df, ncp=0, lower.tail = TRUE, log.p = FALSE) qchisq(p, df, ncp=0, lower.tail = TRUE, log.p = FALSE) rchisq(n, df, ncp=0) all have a slot for the non-centrality parameter "ncp", of the functions for the t and F distributions: dt(x, df, log = FALSE)

There is a problem with `rpois'. It does seem to take care about the order of the arguments. This is an example: > rpois(n=1,lambda=2) [1] 3 > rpois(lambda=2,n=1) [1] 2 0 It obviously uses the first argument as the number of samples to be drawn, which is wrong. I used Version 0.49 Beta (April 23, 1997). Fredrik

There is a problem with `rpois'. It does seem to take care about the order of the arguments. This is an example: > rpois(n=1,lambda=2) [1] 3 > rpois(lambda=2,n=1) [1] 2 0 It obviously uses the first argument as the number of samples to be drawn, which is wrong. I used Version 0.49 Beta (April 23, 1997). Fredrik

Dear List, If any of observed and/or expected data has less than 5 frequencies, then chisq.test (Pearson's Chi-squared Test for Count Data from package:stats) gives warning messages. For example, x<-c(10, 14, 10, 11, 11, 7, 8, 4, 1, 4, 4, 2, 1, 1, 2, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1) y<-c(9.13112391745095, 13.1626482033341, 12.6623267638188, 11.0130706413029, 9.16415925139016,

Sorry, forgot to switch the header to the R group.... --- Globe Trotter <itsme_410 at yahoo.com> wrote: > Date: Thu, 2 Mar 2006 19:35:21 -0800 (PST) > From: Globe Trotter <itsme_410 at yahoo.com> > Subject: Re: [R] calling R's library using C > To: Dirk Eddelbuettel <edd at debian.org> > > Hi, Dirk: > > Thanks for all the help. I thought I would

Is there a method to save a large and complex R object (either as a binary or text file) so that it can be loaded and reused at a later time? Specifically, I am creating large lists (several thousand elements), each element of which is either a vector or a matrix (with ~ 2000 rows). The dimensions of the matrices are not all the same. My ideal would be a set of functions of the form obj

Hi Is there a maximum length for the character string representing a level of a factor? I have a set of several million variables, each a factor of length 19. Each factor level is a character string which in some cases can be many thousands of characters long. I am trying to find out why my analysis fails - I just wanted to rule out the possibility that the internal factor conversion has a