Study On Dynamic Asset Allocation Based On Downside Risk

Asset allocation is the most important part in portfolio investment and decision-making process. In the background of economy globalization, as the financial innovation and financial transactions develop rapidly, there has been a higher-level requirement to asset allocation due to the financial product diversification and more complicated financial risk.Traditional portfolio theory is based on the assumption that investor is risk aversion, that is, investors will choose a less risky portfolio if their expected rates of return is the same. However, the description of risk appetite of investors in traditional portfolio theory is not consistent with actual market conditions; moreover, selection of indicators of risk measurement does not conform to investors'true psychological feelings. In practice of investment decision-making process, only when the portfolio return falls under the target rate of return, does investor consider it the risk of the portfolio. This measurement is called downside risk.We analyze the issue of asset allocation under a downside risk-based framework. First of all, we go through the literature review and theories of investment portfolio in the aspects of the importance of asset allocation, portfolio theories and asset allocation strategies. Then, we compare the mean-lower partial moments (M-LPM) model to the mean-variance (M-V) model. The result proves that M-LPM model has superiority in the application. This is also reflected in the utility function. The traditional expected utility theory can not explain the market anomalies. Prospect theory in behavioral finance brings forward a concept of loss aversion, which indicates that in the utility function of investors,they often give greater weight on negative effect of the loss.The investors are much more sensitive about the loss than the profit. Downside risk measurement and the loss aversion utility reflect investor's attitudes to risk from different points of view.Various types of assets have different performance in different investment cycles; accordingly, the choice of asset allocation strategy is made based on the characteristics of cycles in which the proportion of asset allocation and risk-return balance is considered. The empirical part of this article focuses on our country's security market cycle. Using downside risk as a measurement of risk, we verify the performance of three types of dynamic asset allocation strategy--buy and hold strategy, constant-mix strategy and constant proportion portfolio insurance strategy, during different investment cycle. Empirical results show that buy and hold strategy performs best in the bull market which continues to rise. In a bearish market, constant proportion portfolio insurance strategy is the best choice. Because of its ability to guarantee a minimum insured value, the downside risk can be effectively limited.