Re: st: Hardy-Weinberg with genhw

[Tom Trikalinos <ttrikalin@mac.com>]

At least for the asymptotic test, and the "Understand Why" part of your 
e-mail:
For a biallelic marker (AA;Aa;aa), when -say- the A allele frequency is 
0
the expected numbers of the AA and Aa genotypes are 0.
The asymptotic test for HWE uses a \chi^2 distributed statistic that is 
derived using quantities like:
(observed_AA - expected_AA)^2 / expected_AA
which cannot be defined when expectedd_AA is 0.

I wouldn't use continuity corrections. I don't suppose you can do much.
Perhaps someone else can propose some literature on the subject or a 
workaround.

tom

On Dec 1, 2004, at 11:05 AM, Paolo Grillo, MD wrote:


> I have to perform Hardy-Weinberg equlibrium test with your stata 
> program "genhw" a lot of  gene polimorphism.
> So, I wrote a Stata do.file with the loop "foreach".  Unfortunately 
> the program stopped when an allele frequencies is equal to = 0.
> I have decided to write both to understand why and how to try to work 
> out this problem.
> Best regards
> Paolo Grillo
>
> --
>
>
>
> * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
> Paolo Grillo MD
> Nephrologist Statisticians
> EPOCA Research Center for Occupational, Clinical & Environmental 
> Epidemiology
> Department of Occupational and Environmental Health "Clinica L. Devoto"
> University of Milan
> Via San Barnaba 8
> 20122 Milano - ITALY
> Tel: +39 02 503-20148
> Fax: +39 02 503-20126
> E-Mail paolo.grillo@guest.unimi.it
>
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/