Index-based Strategies and Intraday Stock Dynamics

Abstract

We investigate how index-based strategies affect intraday stock dynamics using a large high-frequency data set of S&P 500 stocks (9.3 billion observations) and estimators that are efficient in the high-frequency setting. Informative intraday patterns are found. In the recent decade, trading volume tilts toward market end; stock correlation increases during the day; cross-sectional beta dispersion decreases. These patterns can be explained by a ``firm-specific open, index-based close'' trading profile, which is based on time-varying informational shocks and trading demands from index-based strategies. We provide further evidence for this mechanism, including a market impact model that generates the observed patterns.

Keywords: market microstructure, index-based investing, intraday stock dynamics, high-frequency estimation, big data

JEL Classification: G11, G12