A Parametric Bootstrap Procedure for Testing Separate Families of Hypotheses.
Technical summary rept.,
WISCONSIN UNIV-MADISON MATHEMATICS RESEARCH CENTER
Pagination or Media Count:
This paper proposes a procedure for testing non-nested families of hypotheses which substitutes raw computing power for asymptotic approximations. Given access to a modern computer, the procedure is practically universally applicable. Examples illustrating its small-sample properties are provided, as are theorems on its asymptotic behavior. Author
- Statistics and Probability