A new measure to study erratic financial behaviors and time-varying dynamics of equity markets

This paper introduces a new framework to quantify distance between finite sets with uncertainty present, where probability distributions determine the locations of individual elements. Combining this with a Bayesian change point detection algorithm, we produce a new measure of similarity between time series with respect to their structural breaks. Next, we apply this to financial data to study the erratic behavior profiles of 19 countries and 11 sectors over the past 20 years. Then, we take a closer examination of individual equities and their behavior surrounding market crises, times when change points are consistently observed. Combining new and existing methods, we study the dynamics of our collection of equities and highlight an increase in equity similarity in recent years, particularly during such crises. Finally, we show that our methodology may provide a new outlook on diversification and risk-reduction during times of extraordinary correlation between assets, where traditional portfolio optimization algorithms encounter difficulties.