Financial correlations at ultra-high frequency: theoretical models and empirical estimation

TL;DR

This paper investigates the correlation between stock returns at ultra-high frequency, comparing empirical data with simple random walk models, and examines how correlations depend on time scales, especially the Epps effect.

Contribution

It introduces a detailed analysis of high-frequency stock return correlations and characterizes stochastic models suitable at very high frequencies, focusing on the Epps effect.

Findings

Correlation dependence on time scales identified

Epps effect characterized at ultra-high frequency

Comparison between empirical data and random walk models

Abstract

A detailed analysis of correlation between stock returns at high frequency is compared with simple models of random walks. We focus in particular on the dependence of correlations on time scales - the so-called Epps effect. This provides a characterization of stochastic models of stock price returns which is appropriate at very high frequency.