An Alternative Nonparametric Specification Test in Autoregressive Conditional Duration Models

Abstract

This paper introduces an alternative testing procedure to test the distribution of the error term in the Autoregressive Conditional Duration (ACD) class of models. In these models, the error term is normally interpreted as the standardized duration by which its probability distribution may have an important impact on the shape of the conditional intensity process of the financial duration in question.This paper illustrates that the testing procedure developed is applicable to various ACD types of models in the literature. Furthermore, when applied to parametric models, such as the Exponential ACD (EACD) and the weibull ACD (WACD), the test can be used as a diagnostic test of the accuracy of the required distributional assumption. Moreover, the hypothetical structure of the test is useful to the specification testing of a number of financial market microstructure hypotheses, especially those related to the information asymmetry in finance. Finally, the testing procedure introduced in this paper differs in many ways from those discussed in existing literatures. This paper shows theoretically and experimentally the statistical validity of the testing procedure, while demonstrating its usefulness and practicality using datasets from New York and Australia Stock Exchange.