```Date: Dec 10, 2012 8:11 PM
Author: Halitsky
Subject: Response to your last re Q and p

You wrote:?But forget S-W and all the others in the list I gave. You want to besensitive to what Stephens (1974) calls "alternative A", so you shoulduse Q = -2*sum{ln p}. Refer Q to the chi-square distribution with df =2*(the # of p's). This is a one-tailed test: the p-value for Q is thearea in the upper tail of the chi-square distribution. For the p'sthat you sent, (Q, p) = (174.954, .00731) for e*u and (254.377, .870e-9) for u^2.?OK, will do.  I?ll compute these before I finish the ?4-ups?.(However, I?d still like to eventually finish the 4-ups as wellbecause for subset=S, method=N, fold=a1, set=1, the plot for u*e notonly hit me right between the eyes, but did so strongly I was almostknocked out.)But one question.For getting p?s back from the t-distribution via the GNU/GSL functionsaccessible from PERL, we wound up doing this:one-tailed p = 1 - ( gsl_cdf_tdist_P( t, df )where gsl_cdf_tdist_P is described here:http://www.gnu.org/software/gsl/manual/html_node/The-t_002ddistribution.htmlSo from what you just wrote re referring Q to the chi-sqauredistribution, I assume that I should do this:one-tailed p = 1 - ( gsl_cdf_chisq_Q( Q, df )where Q and df are computed as you specified and gsl_cdf_chisq_Q isdescribed here:http://www.gnu.org/software/gsl/manual/html_node/The-Chi_002dsquared-Distribution.htmlIf you disagree, please let me know.Thanks very much, as always.
```