A Vulnerability Index for Predicting Extreme Market Events in Hong Kong

Abstract

In this paper, we explore the possibility of developing a “vulnerability” indicator for gauging the health of the economy of Hong Kong. An important measure of “health” to be considered is the popular maximum Lyapunov exponent in the dynamical system literature which measures the sensitivity to initial conditions of a deterministic function. Other key economic and financial indicators that have impact on the Hong Kong financial market such as yield spreads & forward rates are also considered. Lyapunov exponent is often used to indicate the presence of nonlinearity and has not been used as an explanatory variable in the literature, when in fact the Lyapunov exponent also contains useful information about a dynamical system and such information can be usefully exploited. Using stepwise probit regression an indicator of the vulnerability of the financial sector is obtained which is able to indicate empirically observed crisis. An interesting feature of this indicator function is that the local maximum Lyapunov exponent plays a non-negligible role in predicting the status of the economy.