文摘
To forecast exchange rates, this paper proposes a new semiparametric smooth transition autoregressive model by allowing state variables to enter into the transition function in a nonparametric way. We propose a three-stage estimation procedure to estimate both the parametric and nonparametric parts in the new model, and a simulation study is conducted to demonstrate satisfactory finite sample performance. The empirical results, based on the proposed model applied to forecasting five major Asian exchange rates, show that the new model has some advantages in out-of-sample forecasting performance.