Empirical likelihood inference for a class of hysteretic autoregressive models

In this article, we consider empirical likelihood (EL) inference for a class of hysteretic autoregressive models. The main focus of this article is to use the EL method to construct confidence intervals for the parameters and derive the maximum empirical likelihood estimators (MELE) and their asymptotic properties under the conditions that the threshold variable is known or not. Additionally, the testing problem of the nonlinearity of the data is addressed. To illustrate the advantages of solving this model with EL method, we made a simulation study and empirical analysis on the data set of the unemployment rate.