Least Squares Estimation of Linear and Nonlinear ARMAX Models under Data Heterogeneity
In this paper we consider the asymptotic properties of least squares estimators of the parameters of linear and nonlinear ARMAX models under data heterogeneity, where we allow the X-variables to be stochastic time series themselves, possibly depending on lagged dependent variables. These results are obtained by a further elaboration of the results in Bierens [1984, 1987].