Maximum Likelihood Estimation of Dynamic Panel Threshold Model with a Covariate-Dependent Threshold

Abstract

This paper proposes a dynamic panel threshold model with a covariate-dependent and time-varying threshold (DPTCT), in which we allow for the threshold to depend on covariates that vary not only over time but also cross-sectionally. We develop a maximum likelihood estimator of the threshold and slope parameters based on the first difference transformation, and suggest test statistics for threshold effect and threshold constancy. We establish the asymptotic distribution for the threshold estimator and derive the limiting distributions of the suggested test statistics. Monte Carlo simulations are conducted to confirm the asymptotic results and show that the estimation and testing procedures work well in finite samples. The model is illustrated with an application to the inflation-growth nexus.