Non-Invertibility in Some Heteroscedastic Models

Abstract

In order to calculate the unobserved volatility in conditional heteroscedastic time series models, the natural recursive approximation is very often used. Following Straumann and Mikosch (2006), we will call the model invertible if this approximation (based on true parameter vector) converges to the real volatility. Our main result is the necessary condition for invertibility. We will show that the stationary GARCH(p, q) model is always invertible, but certain types of models, such as EGARCH of Nelson (1991) and VGARCH of Engle and Ng (1993) may indeed be non-invertible. Moreover, we will demonstrate it's possible for the pair (true volatility, approximation) to have a non-degenerate stationary distribution. In such cases, the volatility forecast given by the recursive approximation with the true parameter vector is inconsistent.