The Use Of Risk Measurement Tool-CDaR In The Selection Of The Stock Portfolio

One important issue in the field of the financial risk research is Markowitz’s portfolio theory. The development of the portfolio theory includes the mean-variance model, the mean-VaR and the mean-CVaR method. There are lots of mature theories on the studies on VaR (Value at Risk), but what can not be ignored is that most investors are risk averse, meaning that more attention is paid to the loss instead of benefits, so it is necessary to establish a direct loss function associated with the loss of assets, making the maximum benefit under fixed cost. This intuitive and simple way is more in line with the psychological feelings of investors. Following that, three professors from Florida University proposed a new risk measurement tool, namely CDaR. Since it is simple and equipped with many excellent properties, it attracts the attention of many researchers’. Financial risk management has become the forefront of research topics and it develops in a fast speed.This paper introduces the concept and nature of VaR, CVaR and CDaR, showing their advantages and disadvantages and analyzing related processing techniques when CVaR and CDaR are used in decision-making problems. This research focuses on using CDaR (conditional drawdown-at-disk) method in single-parameter risk measurement model, which is to measure the frequency and amplitude of the portfolio’s fall on the particular sample path in a period of time. CDaR expresses a portfolio’s optimization model as a linear programming (LP) problem, achieving am efficient and robust algorithm of the portfolio allocation and making dealing with thousands of optimization problems become possible.