Identifying Common Long-Range Dependence in Volume and Volatility Using High-Frequency Data
22 Pages Posted: 31 Aug 2002
Date Written: August 2002
This paper examines the joint long-run dynamics of trading volume and return volatility in futures contracts on the German stock index DAX using a sample of 5-minute returns and trading volume. Employing robust semiparametric methods of inference on memory parameters, I find that volume and volatility exhibit the same degree of long-memory which is consistent with a mixture-of-distributions (MOD) model in which the latent number of information arrivals follows a long-memory process. However, there is some evidence that volume and volatility are not driven by the same long-memory process suggesting that the MOD model cannot explain the joint long-run dynamics of volatility and volume.
Keywords: Fractional integration, Fractional cointegration, Mixture-of-distributions models, Spectral analysis
JEL Classification: C14, C32
Suggested Citation: Suggested Citation