Combine independent p-values that test the same null hypothesis into an overall summary
combine_independent_p_values.RdFor a vector of p-values \((p_1...p_n)\) that independently test a specific null hypothesis $H_0$, find the relevant quantile of a chi-sq distribution for each p-value, then sum them up. That's a \(\chi^2_n\)-distributed test statistic under $H_0$, and the relevant p-value can be found. Useful for combining analyses after stratification (to deal with potentially complex experimental designs).