Fitting a pth Order Parametric Generalized Linear Autoregressive Multiplicative Error Model

Abstract

This paper is concerned with the problem of fitting a generalized linear model to the conditional mean function of multiplicative error time series models. These models are particularly suited to model nonnegative time series such as the duration between trades at a stock exchange and volume transactions. The proposed test, based on a marked residual empirical process whose marks are suitably defined residuals and which jumps at the estimated indices, is shown to be asymptotically distribution free.