Endogenous Trading Volume and Momentum in Stock-Return Volatility

Abstract

This article examines the ability of volume data to shed light on the source of persistence in stock-return volatility. A mixture model, in which a latent common factor restricts the joint density of volume and returns, is used to relax the assumption of exogenous volume used in previous studies. We use a point-in-time signal-extraction procedure to identify this latent process and a calibrated simulation to conduct analysis of the viability of the model to explain important properties of the data. Using daily returns and volume on individual stocks, our procedure cannot accommodate serial dependence in squared returns.

KEY WORDS:

• Autoregressive conditional heteroscedasticity model
• Mixture model
• Random-effects model
• Small-sample analysis