Econometric analysis of financial trade processes by discrete mixture duration models

We propose a new framework for modelling the time dependence in financial duration processes. The pioneering Autoregressive Conditional Duration (ACD) model introduced by Engle and Russell [1998. Autoregressive conditional duration: a new model for irregularly spaced transaction data. Econometrica 66(5), 1127-1162] will be extended in a way that the duration process is clouded by an unobservable stochastic process. The idea will be put into practice by the Discrete Mixture ACD framework which provides us with a flexible methodology. It will be established by introducing a discrete-valued latent regime variable which can be justified in the light of recent market microstructure theories. The empirical application demonstrates its ability to capture specific characteristics of intraday transaction durations while alternative approaches fail.