Three Approaches to Asymptotic Inference on Nonstationary and Noninvertible Time Series Models

Abstract

Dealing with nonstandard statistics arising from nonstationary and/ or Noninvertible time series models, I discuss how to establish weak convergence results and how to compute limiting distributions. To this end I consider three approaches which I call the eigenvalue approach, the stochastic approach and the Fredholm approach, indicating the advantage and disadvantage of each approach for judicious use.