Three Essays on Equity Market Investment

Abstract

The thesis studies three important topics in equity markets. The first essay examines the relationship between the U.S. and Asia Pacific stock markets, which operate in different time zones. We provide evidence of investors’ asymmetric reactions to good and bad news. Our results suggest that extreme positive returns in the U.S. market have a lower impact on Asia Pacific stock markets than normal positive returns. We find a significant asymmetric effect for surprise returns, but not for ordinary returns. The second essay investigates the price discovery process of Asia Pacific country ETFs. The price of a country ETF is influenced not only by net asset value but also by information during U.S. trading hours. In this study, I examine six Asia Pacific ETFs from 2006 to 2020, using linear regression and tree-based ensemble methods. The results indicate that local market trading hours significantly affect the predictive power of ETFs and the S&P 500 Index. ETF returns reflect short-term expectations of underlying indices, not just reactions to the U.S. market. The third essay examines the accuracy of target price forecasts by sell-side analysts, using machine learning techniques. We analyze forecasts for U.S. listed companies from 1999 to 2021, incorporating market, firm, and analyst-level variables to predict target price accuracy. Our ensemble methods predict target price errors and achievements effectively. Long-short portfolios based on these predictions outperform benchmarks in cumulative return and Sharpe ratio.