Answers are grounded in this paper and cite their sources.
E2= fiF* f2 +f1 F* it was studied by Wishart (1932) and by Tang (1938) who gave a few values of the probability integral. Patnaik (1949) approximated to the distribution of F* by a central F distribution. It is clear that any tabulation of the distribution of F* will be a formidable proceeding. There are three parameters fl, f2 and A which intervene explicitly. If we are concemed with the power function of the F-test there must also be the values of the central F at which p(F*) will need to be evaluated. We discuss here the log-transformation of F* and the possibility of a normal distribution as an approximating function to the transformed variable. ?
No takes yet. Share an insight, caveat, or question.
Barton et al. (1960) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: