A1193
Title: Simulation-based multiple testing for many non-nested multivariate models
Authors: Florian Richard - Universite Laval (Canada) [presenting]
Lynda Khalaf - Carleton (Canada)
Abstract: A multivariate extension of exact specification tests for non-nested models is proposed. The test is finite-sample exact under the assumption of Gaussian errors and is easily generalized to a multiple-model hypothesis via a combined alternative. Valid inference results are obtained using bootstrapped Monte Carlo p-values, even when the distribution under the null hypothesis is intractable. Both Gaussian and non-Gaussian error structures are considered through bootstrapping, and it is shown that the test possesses good size and power properties via simulations. Finally, empirical applications to asset pricing are presented by testing benchmark factor models against single and multiple alternatives.
