Relation between discontinuous variation and stock return movement based on poisson process: Evidence from China

Abstract

This study constructs an analytical framework grounded in the principles of Poisson distribution and the process of discontinuous variation. Through empirical examinations conducted on all publicly listed firms in Chinese stock markets, the research uncovers an inverse relationship between the discontinuous variance of stock returns from the preceding month and the stock returns of the subsequent month. Statistically, the average monthly return of the portfolio with the lowest discontinuous variation is 1.32%, and the average monthly return of the portfolio with the highest discontinuous variation is 0.30%, indicating that an increase in discontinuous variance by 1 unit is associated with a subsequent decrease in stock returns of 1%, on average. Furthermore, portfolios engaged in long-short arbitrage, which are formulated based on sorting portfolios of different discontinuous variances, exhibit superior performance compared to the market portfolio. The effective annual return of this long-short portfolio is 1.44%, which is larger than that of the market index (1.19%). Also, the Sharpe ratio of the long-short portfolio is 0.19, which is almost double that of the market index (0.10). The results of portfolio performance suggest that trading strategies based on Poisson distribution and discontinuous variation sorting possess a significant capacity to generate abnormal returns.