A1961
Title: Finite-sample properties of model specification tests for multivariate dynamic regression models
Authors: Koichiro Moriya - Keio University (Japan) [presenting]
Akihiko Noda - Meiji University (Japan)
Abstract: A new model specification test is proposed for multiple-equation systems with cross-equation error and dynamic regressor-error dependences. Conventional tests often rely on exogeneity conditions strong enough to ensure consistency of the OLS estimator. These exogeneity conditions are violated when regressors and errors are dynamically dependent, rendering conventional model specification tests invalid. To address these limitations, the relationship among alternative exogeneity conditions is clarified, the consistency of competing multiple-equation estimators is characterized, and a generalized Durbin estimator is proposed for multiple-equation systems with an intercept, cross-equation error and regressor-error dependences. The estimator remains consistent under the weakest exogeneity condition. Its asymptotic distribution is derived and Wald tests are constructed. Monte Carlo experiments confirm that the bootstrap-based Wald test substantially improves finite-sample size control. An application of the bootstrap-based Wald test to the Fama--French multifactor models leaves the null hypothesis unrejected in cases where competing FGLS-based tests reject it.
