IDENTIFICATION AND ANALYSIS OF STOCK MARKET DYNAMICS

Abstract

This thesis investigates the stock market dynamics of long run cycles and short run fluctuations. First, based on a system adaptation framework, we employ a wavelet multi-resolution analysis to decompose the stock price signal and retrieve its informative frequency components. Using the recovered signals, an empirical index is developed to forecast the stock market turning points. This index is successfully applied to US, UK and China markets, where all major turning points are well forecasted. Second, to explore the fast market dynamics, we conduct an empirical study to investigate the transient intermarket reactions under 9/11 terrorist attack. The results reveal the time-varying spillover effects and co-movements of inter-markets. Third, the emerging stock markets are usually characterized as having high volatility with low predictability. Selecting the China stock market as a representative of the emerging markets, we carry a comprehensive investigation on the market driving forces behind such volatile characteristics.