Jump Intervals of Stock Price Have Power-Law Distribution: An Empirical Study

Abstract

Taking the power-law behavior of human activities into consideration, we conduct an empirical study on the distribution of jump intervals after using BNS nonparametric method to detect jumps in 5 min closing data of HIS. Our result shows that there is a “power law” in jump intervals, and Fokker-Planck distribution is the more suitable distribution to describe jump intervals than the traditional Poisson process. So the jump diffusion model of power law can depict the movement of stock price more accurately.