R devel - Apr 2005 - fisher.g.test.single cannot handle time series longer than 341 (PR#7782)

Full_Name: Benjamin Tyner
Version: R version 2.0.1, 2004-11-15
OS: Linux (kernel version 2.6.10-1.770_FC3)
Submission from: (NULL) (4.64.8.220)


On my build, for example:

library(GeneTS)
fisher.g.test.single(rnorm(341)) #this works fine
fisher.g.test.single(rnorm(342)) #this fails with the following error:

"Error in if (pval > 1) pval <- 1 : missing value where TRUE/FALSE
needed"

The error always occurs for n>=342. This also affects the behavior of
fisher.g.test. I suspect the problem may lie in periodogram.spec

The FAQ says:

    Bug reports on contributed packages should be sent first to the package
    maintainer, and only submitted to the R-bugs repository by package
    maintainers, mentioning the package in the subject line.

You have not mentioned the package in the subject line, and you are not 
the stated maintainer. The report on R-bugs has been closed.

On Fri, 8 Apr 2005 btyner@stat.purdue.edu wrote:
> Full_Name: Benjamin Tyner
> Version: R version 2.0.1, 2004-11-15
> OS: Linux (kernel version 2.6.10-1.770_FC3)
> Submission from: (NULL) (4.64.8.220)
>
>
> On my build, for example:
>
> library(GeneTS)
> fisher.g.test.single(rnorm(341)) #this works fine
> fisher.g.test.single(rnorm(342)) #this fails with the following error:
>
> "Error in if (pval > 1) pval <- 1 : missing value where
TRUE/FALSE needed"
>
> The error always occurs for n>=342. This also affects the behavior of
> fisher.g.test. I suspect the problem may lie in periodogram.spec
>
> ______________________________________________
> R-devel@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
>
-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
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, UK                Fax:  +44 1865 272595