Volatility and autocorrelation in European futures markets

Purpose - This paper seeks to investigate the relationship between volatility and autocorrelation in major European stock index futures markets. Design/methodology/approach - The methodology is based on the exponential autoregressive model with conditionally heteroskedastic errors (EAR-GARCH). Findings - The evidence points to a negative relationship between volatility and autocorrelation. Specifically, autocorrelation is low during volatile periods and high during calm periods. This evidence is in agreement with LeBaron's findings for US stock market returns, suggesting that return dynamics are similar across asset categories. Research limitations/implications - An obvious limitation of this study is the lack of a theoretical justification for the observed relationships in futures markets, an area where future research should be directed. Practical implications - The observed relationships suggest that futures prices are non-linearly predictable so that short-term trading could produce abnormal returns. Originality/value - The paper documents a negative relationship between volatility and autocorrelation in major European futures markets. This finding should be of interest to researchers and market participants.