Uncovering Linkages in International Financial Markets Via Boosting

Abstract

Boosting is a powerful learning idea that has produced dramatic improvements in predictive learners' performance. In this paper, we use a Gradient Boosting Machine to discover linkages in three main stock market indices in the world during periods of large positive price changes. The uncovered lead-lag patterns reveal that the Standard & Poors 500 index contains incremental information that could be used to enhance the predictability of the large positive returns in the three main stock market indices in the world. This result cannot be explained by neither FTSE 100's nor Nikkei 225's information content.

Keywords: International stock markets, Causality

JEL Classification: G15, C32