HOME Recent Events RMI in the News ALUMNI
  Issue 18 | Archive February 2014

Volume and Volatility in a Common Factor Mixture of Distributions Model

A paper by Xiaojun He (Syracuse University) and Raja Velu (Syracuse University)

RMI¡¯s visiting professor Prof. Raja Velu from Syracuse University¡¯s Whitman School of Management, in collaboration with his PhD student Xiaojun He, has written a research paper entitled ¡°Volume and Volatility in a Common Factor Mixture of Distributions Model.¡± This paper appeared in the Journal of Financial and Quantitative Analysis in December 2013.

In the paper, the authors argued that the relationship between a security¡¯s return and its trading volume is of great interest to financial economists. In the so-called Mixture Distribution Hypothesis (MDH) model, it is postulated that the return and volume are jointly related to an unobservable dynamic information-flow variable and are equilibrium outcomes of the information impact.

However, many studies show evidence unfavourable to the standard MDH model. He and Velu conjectured that the model is not wholly successful because it treats securities in multi-asset financial markets as isolated from each other. They then introduced a common factor MDH model that could account for market-wide movement of related securities. The factors may include changes in monetary policy, tax regime, political events, etc. The model also incorporated the features of volatility dynamics. For empirical analysis, He and Velu used half-hour intraday data for the securities included in Dow Jones and S&P composites.

The results are generally consistent with their predictions. Moreover, the market-wide analysis on reduction in the persistence of volatility indicates the model performs better for securities with large capitalization and small trading volume.

The results of this study are useful for pricing of a security and will also aid regulators, exchanges and other participants to improve market design, especially episodes of domestic and international financial turmoil. On the other hand, the authors cautioned that the model is unable to differentiate the informational effect from the impact of uninformed trader¡¯s behaviours on the cross-security variation and explained that this issue will be explored in future research.

Back To Newsletter

Published quarterly by Risk Management Institute, NUS
Editor: Ivy Wang (rmiwy@nus.edu.sg)