On randomised multiple comparison procedures for discrete distributions (Joint work with Alex Lewin)
by Dr Elena Kulinskaya
Abstract: Randomised multiple comparisons procedures are introduced as a solution to the problem of multiple comparisons for discrete test
statistics. The classical concept of randomised tests is extended to multiple
comparisons. The critical function of the randomised p-values is proposed as a measure of evidence against the null hypotheses.
This approach makes all theory of multiple comparisons
developed for continuously distributed statistics automatically applicable to the discrete case. Examples of both FWER (Bonferroni(1935)) and FDR (Benjamini-Hochberg (1995)) procedures are discussed.
For More Information: Dr Owen Jones: firstname.lastname@example.org