Market Crash Prediction Model for Markets in A Rational Bubble

TL;DR

This paper introduces a new white box model for predicting market crashes based on rational bubble assumptions, demonstrating successful predictions on Dow Jones and Bitcoin markets, highlighting its sensitivity and generalization.

Contribution

The paper proposes a novel white box model for crash prediction rooted in rational bubble theory, expanding beyond traditional methods like LPPL.

Findings

Successfully predicted major crashes in Dow Jones and Bitcoin markets

Model shows high sensitivity to market signals

Demonstrates good generalization across different market types

Abstract

Renowned method of log-periodic power law(LPPL) is one of the few ways that a financial market crash could be predicted. Alongside with LPPL, this paper propose a novel method of stock market crash using white box model derived from simple assumptions about the state of rational bubble. By applying this model to Dow Jones Index and Bitcoin market price data, it is shown that the model successfully predicts some major crashes of both markets, implying the high sensitivity and generalization abilities of the model.