Thank you very much to Sundar Dorai-Raj, Fernando TUSELL and Thomas Hotz for
their quit and useful answers to my question (see below).
By the way, some of them asked me "what's the practical difference
between 0.999999999 and 1.0?" ANSWER: My interest on getting exact p-values
lies in the fact that for some simple meta-analyses (e.g., Fisher Test for
Combining Probabilities) you need to input the exact p-values of an array of
tests to perform an "overall" statistical test. Further, for both the
standard Bonferroni correction and the sequential Bonferroni correction you
need to input exact p-values.
MY QUESTION:
I want to get the exact p-values, on 1 degree of freedom, for an array
of chi-square values. When my chi-square values are equal or lower than
29.7, I get the exact associated p-values. Thus, for instance:
>pchisq(29.7, df=1)
[1] 0.9999999
However, when my chi-square values are greater or equal to 29.8 what I get
is:
>pchisq(29.8, df=1)
[1] 1
Could anyone tell me how to fix this trivial issue? Very grateful, Jose
M. Fedriani
THEIR ANSWERS (SUMMARIZED):
Increase the 'digits' option:
> options(digits = 17) # the max
> pchisq(29.8, df=1)
[1] 0.99999995210080161
On the other hand, they also caution that the "options(digits=...)"
merely affects the presentation of p-values; thus, the accuracy of those
"extended values" could be questionable.
|