Comparison of time-series of risk aversion between Singapore, Hong Kong and USA
Chua, Yi Ting
Tan, Peggan Peck Hia
Date of Issue2007
College of Business (Nanyang Business School)
This paper aims to estimate and compare the time series of risk aversion for Singapore, Hong Kong and United States of America (USA). Like Chou, Engle and Kane, (1992), or CEK, we used the time-varying parameter (TVP) GARCH-M model to calculate the historical series of risk aversion parameter. For comparison purpose, we also use the conventional ratio formula with and without rolling sample as well as rolling sample GARCH-M model. Results from TVP technique suggest that investors in USA stock market are the most risk averse. This is followed by Hong Kong and Singapore which is an interesting finding. It is because Hong Kong investors are often perceived to be more risk-taking than Singapore investors. This perception may be based on the actions taken by people in Hong Kong rather than their inherent preference to risk. It could be true that residents in Hong Kong take more risk not because they want to, but because of the need to or are forced to. In addition, our results also show that the risk aversion time series for Singapore and Hong Kong are relatively more correlated (0.821), while the risk aversion in Singapore and Hong Kong are less correlated with the risk aversion in USA respectively (0.611 and 0.689). We also analyzed the historical volatilities for all three economies. We found that while economic events influence fluctuations in volatility, risk aversion over time remained constant which implies that risk aversion is an innate nature and is not affected by the events.
Final Year Project (FYP)
Nanyang Technological University