Abstract: The paper considers the so-called double AR() model,
where i.i.d. . It is shown that the necessary and sufficient condition for the existence of a strictly stationary solution to the model is that the top Lyapounov exponent , defined in the paper, be negative; the solution is then unique and geometrically ergodic. The necessary and sufficient condition for the existence of a strictly stationary solution to the model with is also obtained. The maximum likelihood estimator of the parameters in the model is shown to be asymptotically normal. The condition for this is again only that is negative which includes the case with some roots of on or outside the unit circle, and the case with . The result is novel because all kinds of estimated 's in these cases are not asymptotically normal in the classical AR() model with i.i.d. errors; it may provide new insights in this direction.
Key words and phrases: Asymptotic normality, double autoregressive model, maximum likelihood estimator, stationarity, geometric ergodicity.