The Econometrics of Financial Duration Modeling

TL;DR

This paper investigates the estimation and inference challenges in financial duration models, revealing that traditional estimators may fail or behave non-standardly when durations have heavy tails or infinite mean, especially in high-frequency trading data.

Contribution

It provides new theoretical insights into the asymptotic behavior of likelihood estimators in duration models with heavy-tailed data, extending understanding beyond classical assumptions.

Findings

Likelihood estimators are sensitive to tail behavior of durations.

Asymptotic normality fails for durations with tail index less than one.

Estimators exhibit mixed Gaussian behavior with non-standard convergence rates.

Abstract

We establish new results for estimation and inference in financial durations models, where events are observed over a given time span, such as a trading day, or a week. For the classical autoregressive conditional duration (ACD) models by Engle and Russell (1998, Econometrica 66, 1127-1162), we show that the large sample behavior of likelihood estimators is highly sensitive to the tail behavior of the financial durations. In particular, even under stationarity, asymptotic normality breaks down for tail indices smaller than one or, equivalently, when the clustering behaviour of the observed events is such that the unconditional distribution of the durations has no finite mean. Instead, we find that estimators are mixed Gaussian and have non-standard rates of convergence. The results are based on exploiting the crucial fact that for duration data the number of observations within any given time span is random. Our results apply to general econometric models where the number of observed events is random.