On the critical value behavior of multiple decision procedures (Q2742767)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: On the critical value behavior of multiple decision procedures |
scientific article; zbMATH DE number 1650392
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | On the critical value behavior of multiple decision procedures |
scientific article; zbMATH DE number 1650392 |
Statements
23 September 2001
0 references
Bonferroni test procedure
0 references
hazard rate
0 references
independent p-values
0 references
multiple comparisons
0 references
multiple test procedure
0 references
step-down test
0 references
On the critical value behavior of multiple decision procedures (English)
0 references
The authors investigate the critical value behaviour of several multiple decision procedures when the number \(n\) of parameters or hypotheses under consideration increases. The authors obtain results of the following type. Suppose that the sequences of critical values to be compared are denoted by \(v_n\) and \(w_n.\) Since it is almost impossible to give explicit formulas for the critical values, and much less for the differences of these, the authors first derive asymptotic expansions for the critical points by means of a special method. In the majority of cases, these asymptotics do not provide enough information to yield an appropriate asymptotic expansion for the differences of the \(w_n\) and \(v_n\) by simple substraction. The authors employ the knowledge of the asymptotic behaviour of the corresponding differences based on uniformly distributed observations such as, for example, \(p\)-values to obtain a sequence \(g_n\) such that \(\lim_{n \to \infty} g_n(w_n - v_n) =g\) for some positive constant \(g.\)
0 references