Extraction of Information Content Exchange in Financial Markets by an Entropy Analysis

Abstract

Recently, there has been an explosive interest in the literature about modeling and forecasting volatility in financial markets. Many researches have focused on energy markets and oil volatility index (OVX). In this article, we aim first at showing if there is an exchange of information between two stock time series, and then at evaluating what is the direction of this information flow. In particular, we propose an entropy-based approach that exploits two objective metrics, namely Mutual Information (MI) and Transfer Entropy (TE), that does not require a parametric model and is directly applicable on the data. The experimental outcomes, applied on Brent and WTI crude oil prices and their volatility index for the period from May 10, 2007 till July 03, 2018, demonstrate the effectiveness of the proposed method.