Choosing an optimal method to combine P-values.

摘要: Fisher (1925) was the first to suggest a method of combining p-values obtained from several statistics and many other methods have been proposed since then. However, there is no agreement about what best method. Motivated by situation that now often arises in genetic epidemiology, we consider problem when it possible define simple alternative hypothesis interest for which expected effect size each test statistic known determine most powerful this hypothesis. Based on method, show information sizes can be used obtain weights Liptak's p-values. We present extensive simulation results comparing illustrate real example epidemiology how deduced.