The British Stock Market, currencies, brexit, and media sentiments: A big data analysis

Abstract

In this study, we use a machine learning framework and draw on an extensive body of media articles on Brexit to provide evidence of cointegration and causality between the sentiments of the media and the movement of British currency. Our contribution to the literature is novel. In addition to applying lexicons commonly used in sentiment analysis, we devise a method using Bayesian learning to create a more context-aware and informative lexicon for Brexit. By leveraging and extending this method, we reveal the relationship between original media sentiment and related economic and financial variables. Our method is a distinct improvement over existing methods and can better predict out-of-sample outcomes than conventional ones.