The Strong Mixing Data-Based Semiparametric Smooth Transition Regression Model

In this paper, we generalize the semiparametric smooth transition regression model proposed by Wang (2012a), to adapt for the strictly stationary strong mixing data and strong mixing data with deterministic trends. The unknown bounded smooth function embedded in the smooth transition function is estimated by series estimator, the consistency and asymptotic normality properties of estimators are proved employing nonlinear least square regression theory and series estimator approach. Variance matrix estimation and hypothesis testing problems are also discussed based on estimated standard errors. The new model is then used to study the annually inflation rates of China.