[R] partitioning chi-square statistic (g squared)

[Peter Ehlers]

christopher compeau wrote:
> hi all -
> 
> is there a package or library that contains a function for partitioning the
> chi-square statistic of an I X J contingency table into its respective
> independent parts?

Do you mean the chisquare contribution from each cell in the table?
If so, just square the 'residuals' component of the output:

  x <- sample(5:20, 12, replace=TRUE)
  m <- matrix(x, nr=3)
  chtest <- chisq.test(m)
  residuals(chtest)^2

See the 'Value' section of ?chisq.test.

If you mean the "G-test" (Wikipedia), you can extract the
'observed' and 'expected' components of the chisq.test output
to calculate G. Use the chisquare distribution to obtain a
p-value.

  -Peter Ehlers

> 
> i looked around for this, but i didn't find anything. perhaps there's
> another name for this sort of analysis? i know it as "g-squared".
> 
> thanks,
> chris.
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
>