New volatility evolution model after extreme events

TL;DR

This paper introduces a new two-stage model for stock market volatility after extreme events, showing a transition from stretched exponential to power law decay, supported by empirical data and investor behavior hypotheses.

Contribution

The paper proposes a novel two-stage volatility evolution model and validates it with high-frequency data and behavioral hypotheses, advancing understanding of post-extreme event market dynamics.

Findings

Volatility follows a stretched exponential decay initially.

Volatility transitions to a power law decay later.

Investor behavior shifts from uninformed to informed states.

Abstract

In this paper, we propose a new dynamical model to study the two-stage volatility evolution of stock market index after extreme events, and find that the volatility after extreme events follows a stretched exponential decay in the initial stage and becomes a power law decay at later times by using high frequency minute data. Empirical study of the evolutionary behaviors of volatility after endogenous and exogenous events further demonstrates the descriptive power of our new model. To further explore the underlying mechanisms of volatility evolution, we introduce the sequential arrival of information hypothesis (SAIH) and the mixture of distribution hypothesis (MDH) to test the two-stage assumption, and find that investors transform from the uninformed state to the informed state in the first stage and informed investors subsequently dominate in the second stage. The testing results offer a supporting explanation for the validity of our new model and the fitted values of relevant parameters.