Two Misspecification Tests for the Simple Switching Regressions Disequilibrium Model

Two specification tests for switching regimes disequilibrium models are developed. The first is an asymptotically locally optimal Lagrange multiplier test of endogeneity of a set of regressors, which takes the convenient form of a LM significance-test of certain regression residuals. The second is a Hausman specification test of the accuracy of regime classification information.