An Improved Stochastic Conditional Duration Model

摘要

This paper proposes an improved version of stochastic conditional duration model, which is used to analyze the dynamics of the intra-daily trading activity of financial markets, and a novel estimation methodology for the model. The model imposes mixtures of bivariate normal distribution family on the innovations of the observation and latent equations of the duration process. This extension allows the model not only to capture the asymmetric behavior of the expected duration but also to easily accommodate a rich set of dependence structures between the two innovations. A simple Monte Carlo study is conducted to assess the performance of the proposed model and estimation method, and an empirical illustration of the model and estimation method is provided with the IBM transaction data.