| Since"Interim Measures for the Administration of Securities Investment Funds" enacted in November 1997, China's fund industry has grown into one of the most important institutional investors after more than ten years'searching for the development of securities markets. Securities investment funds grow geometrically both in quantity and in size and at the same time, the fund species are also getting richer. Therefore, how to analysis the value of investing in funds in accordance with the risk of product should be the primary problems faced by investors in such a complex investment environment. The most direct and effective way is to score the funds and give them a classification based on the performance of funds in the past. In this way investors can select a fund as an ideal investment according to their own preferences. The "scoring" process is right the fund performance evaluation process.The financial crisis sweeping through the world gives us a profound lesson, so I believe that the outbreak of the financial crisis is rooted in the weak awareness of investors towards financial risk, as well asthe much too optimistic attitude when judging the development of the market. The result is that, they pay excessive attention on the earnings performance. It can be a safe conclusion that the ways to reduce risks and increase the income are the two core issues which investors and academics have always concerned. The two issues are closely related, but contradictory to each other as well. In this article, therefore, I will always focus on these two investment options of the Fund which have a decisive role in the analysis of the Fund chosen and use the two elements as the basis for performance evaluation. Inspiration by the financial crisis, I will primarily focus on how to assess the risks scientifically so as to discover and effectively circumvent the risks.Therefore, in the process of modeling, I first assumed that all the investors in the market are risk aversion, and then use a number of ways to measure the risk. There are three models which are built based on the idea of Giorgi (2005), who put forward "reward—risk pair" for the first time.In this paper, first I will make a brief review of the development of China's fund industry, and give an objective description about the actuality of fund industry in China. Secondly, I will introduce the traditional evaluation methods for Fund in accordance with development sequence of the theories, including the three classic indicators in funds performance evaluation which come from investment portfolio theory, as well as the application of data envelopment analysis (DEA) theory in fund performance evaluation. In the third part of this article, I will describe in more details about the data envelopment analysis based on the second-order stochastic dominant method. Through an analysis of previous studies, six DEA models are suitable for the evaluation of fund performance, which are the models used to evaluate Fund performance in this paper. The fourth part of this paper will do empirical study and chose different models for investors of different risk preferences according to specific market conditions during the period in order to assess fund performance. The last part of the paper is the conclusion.This paper chose the method of DEA to do the research on fund performance problem is primarily due to the traditional valuation methodscontain so many constraints. First, the premise of market validity.China's securities market is still at the developing stage, not yet a mature and effective operation and the monitoring system, the market information can not be completely open, fair and non-blocking communicated, so China's securities market is still in a state of invalid or weakly efficient; second, the traditional methods of fund performance evaluation highly depend on the effectiveness of the market portfolio selection. Finally, the traditional methods of fund performance evaluation should also assume that there is a kind of risk-free interest rate, but in real life there is no risk-free assets and risk-free interest rate exists, then how to choose a risk-free interest rate on evaluatingresults becomes another reason for the distortion. Since during the evaluation, the low-risk fund will often be considered at a higher level, and the high-risk fund will be access to a lower performance. The fund performance evaluation based on DEA models not only overcome these fatal flaws, but also study some certain factors affecting the results of the evaluation. In addition, the Fund with the traditional methods of performance evaluation, DEA method does not need assumptions of any function form between input and output elements, only to meet the basic DEA model assumptions: the production set is (efficient frontier surface) of the convex , and the smallest of invalidity; parameter estimation is compared with the average efficiency performance, and DEA methods (non-parametric methods) is measured relative efficiency of Pareto efficient frontier, DEA can found not only the poor efficiency of the Fund , but can estimate the degree of invalidity, as well as the way to improve the funds efficiency; DEA method can inspect the relative importance of elements, you can check (different transaction costs or different risk measure) the impact of the marginal contribution to the income of the fund with each input element, in order to obtain the optimal combination of income and portfolio allocation of resources inherent in the meaning of measurement; DEA method can also be used to study the size of the fund performance of the Fund and so on.To sum up, through the fund performance data envelopment analysis we can obtain amount of information about the acts, attributes and the relationship between fund performance that the traditional methods of fund performance evaluation can not find.In accordance with different risk preferences, six models used in empirical study in the paper can be divided into risk -reward DEA model and security - reward DEA model. Risk - reward DEA model use a measure of risk as the input elements and the mean return as the output elements; security - proceeds DEA model use constant as input elements and return measure and safety measrue as output elements. Therefore, this model is a pure DEA model of output. The empirical results from this paper show that, all the security - reward models are DEA models with risk aversion, thus the investors sensitive for the risk can make use of such models for fund performance evaluation. Risk -reward DEA models can be classified into different types according to the degree of risk aversion and expectations for the reward, I divide them into three different types of models according to their style: the risk preference model,the risk absolutely aversion model and the normal risk aversion model. I believe that the empirical model in this paper has a strong practical value, and it can be used during the investment as a reference when making investment decision for investors. |
PDF Full Text Request