To lag or not to lag? How to compare indices of stock markets that operate at different times

TL;DR

This paper investigates how to compare stock market indices operating at different times by considering both original and lagged data, improving understanding of their interrelations.

Contribution

It introduces a method to incorporate both original and lagged indices in correlation analysis, enhancing insights into global market relationships.

Findings

Including lagged indices improves correlation network understanding.

Original and lagged indices together reveal more about market interdependencies.

The approach aids in better modeling of asynchronous market behaviors.

Abstract

Financial markets worldwide do not have the same working hours. As a consequence, the study of correlation or causality between financial market indices becomes dependent on wether we should consider in computations of correlation matrices all indices in the same day or lagged indices. The answer this article proposes is that we should consider both. In this work, we use 79 indices of a diversity of stock markets across the world in order to study their correlation structure, and discover that representing in the same network original and lagged indices, we obtain a better understanding of how indices that operate at different hours relate to each other.