On maximum likelihood estimators for a threshold autoregression (Q1299006)

The SETAR (self-exciting threshold autoregressive) model is defined by \(X_i= h({\mathbf X}_{i-1}, \theta)+ \varepsilon_i\), \(i\geq 1\), where \({\mathbf X}_{i-1}= (X_{i-1},\dots, X_{i-p})'\) is observable, \(\theta\in \Re^{p-1}\) are parameters. For a stationary SETAR model, \textit{K.S. Chan} [Ann. Stat. 21, No. 1, 520-533 (1993; Zbl 0786.62089)] obtained the consistency and limiting distribution of the least-squares estimator for the underlying parameters. In this paper, the author derives similar results for the maximum likelihood estimators of the same model, under some regularity conditions on the error density, not necessarily Gaussian.