Title: On resampling tests and the nested simulation problem Authors:  Daniel Gaigall - FH Aachen University of Applied Sciences (Germany) [presenting]
Julian Gerstenberg - Goethe University Frankfurt (Germany)
Abstract: Statistical tests based on resampling procedures are considered. A general framework is introduced that covers, in particular, bootstrap and permutation techniques for the computation of approximate quantiles as critical values in model specification testing. For the investigation of properties of such tests, Monte-Carlo simulation studies are customary. The resampling procedure leads to a nested simulation and ultimately to a nested simulation estimator for the rejection probability of the test. Choosing both the number of replications and the size of the simulation study large results in a considerable computational effort. To circumvent this problem, the so-called warp-speed method has become popular recently. For that reason, the related warp-speed estimator is revisited. Besides, the latest results for the nested simulation estimator indicate that a moderate or even rather small number of replications is sufficient to obtain useful simulation results. This enables a substantial reduction of the computational effort.