The Studies On Performance Evaluation Of Chinese Security Funds

During seven years'development of Chinese security fund industry, under the increasing standardizing and supporting of administration authorities, the funds'scale is increasingly expanding, the products are increasingly innovating, the regulation is being progressively perfected, market participators, especially institutional and individual investors, such as insurance funds, social security fund, enterprise annual fund, etc. are growing up rapidly, and thus Chinese fund industry is developing rapidly. The performance evaluation of Chinese fund attracts attention of market participators, but the theoretical studies about it are insufficient in China now which doesn't suit to the urgent demand on rational evaluation of Chinese fund, so we need supplement and develop the research on the performance evaluation of Chinese fund urgently. The fund performance evaluation mainly includes the evaluation of overall performance, the ability of securities selecting and market timing, performance persistence and comprehensive performance of the funds. At present domestic studies focus on evaluating the overall performance of Chinese funds utilizing foreign methods, but the studies on the ability of securities selecting and market timing, performance persistence and comprehensive performance evaluation are not enough and deficient. The defects mainly include misapplication of foreign methods, the research on the sensitiveness of performance evaluation to different market portfolio benchmarks and comprehensive performance evaluation. This paper arranges the inherent logic framework of the studies on the fund performance evaluation, and sets up more perfect system of the fund performance evaluation. On the basis this paper carries out overall and systematic researches on the overall performance, the ability of securities selecting and market timing, performance persistence and comprehensive performance evaluation by means of the methods from portions to whole which expands the present studies'breadth and depth. This paper is organized as follows: Chapter 1 introduces the cause of choosing this subject, structure and main innovations. Chapter 2 summarizes exiting studies on fund performance evaluation in China and foreign countries. Firstly, this paper comprehensively reviews theoretic restudies on fund performance evaluation and the development of appraisement institutions in western countries. Then this paper summarizes the theoretic and empirical studies at home. Chapter 3 carries out further theoretic and empirical studies on the overall performance evaluation of the funds. Domestic studies mostly introduce foreign risk adjusted methods (mainly classical methods) and simply apply them to Chinese funds'empirical studies, and the uses of classical methods in many studies are improper or wrong and thus the empirical results drawn form those studies are biased. This paper carries out the through and painstaking analysis of the assumption and suitability of various kinds of models and methods, summarizes the problems of existing studies, and points out the mistakes caused by all kinds of methods especially classical methods when them are misused. On the method research, this paper introduces the method produced by Bird and Gallagher in 2002 for the first time that bases on investor's preference and adopts the prior four moments of active return to evaluate the fund performance, which is called Moments Method based on Investors'Preference in this paper. They give the weights of the four moments subjectively, and provide with the quintiles for the four moments. This paper considers this method to be irrational and suggests an improved method that carries out the standardizing of the four moments and gives the weights of the four moments through Analytical Hierarchy Process. Then this paper uses the classical methods (including Sharpe Ratio, Treynor Index and Jensen Index), the improved risk adjusted methods (including information ratio, M 2 measure, performance index and morningstar rating), the multifactor model (including three-factors and dual-factors model) and moments method based on investors'preference to carry out the empirical studies on the overall performance of 54 closed-end funds and 12 opened-end funds from Nov. 1, 2002 to Nov. 5, 2004. There are four characteristics in the research on the overall performance evaluation of Chinese funds in this paper. First of all, some methods (Jensen Index, Treynor Index and multifactor models) depend on the choice of different market portfolio benchmarks. Domestic studies mostly draw the empirical conclusion on the basis of single-factor index at present. This paper carries on empirical studies on the sensitiveness of Jensen Index and Treynor Index to different market benchmarks adopting eight indexes and evaluates the consistency of the results. Empirical results indicate that Jensen Index has very low sensitiveness to different indexes, and Treynor Index's sensitiveness is slightly higher.The influences of different indexes as benchmarks on the empirical results are very limited. Secondly, on the basis of three-factor model and the statistics significance of the independent variables, this paper puts forward the dual-factor model for the first time in China, and applies it to the performance evaluation of Chinese funds. Compared with three-factor model, the explanation ability of the dual-factor model improves significantly and it has more rationality. Thirdly, the empirical results indicate that moments method based on investors'preference is efficient in evaluating the fund performance. The improved method of this paper is more rational than that of Bird and Gallagher, which is a breakthrough in the existing researches on the fund performance evaluation. Lastly, in regard to the sample funds, domestic studies mostly base on closed-end funds. This paper selects 54 closed-end and 12 opened-end funds as samples, carries on comparative researches of opened-end and closed-end funds, and thus learns two kinds of funds'performance on the whole at present. Finally this paper carries on consistency inspection to the empirical results of different methods. Results show that there isn't obvious evidence that indicates Chinese funds can defeat the market, and only a few funds have the ability to defeat the market evidently. In terms of the behaviors of two kinds of funds on the whole, the opened-end funds obviously superior to the closed-end funds, but as regards specific funds it is mainly closed-end funds that are most outstanding and worst. Chapter 4 studies the ability of securities selecting and market timing. Firstly, this paper expands TM, HM and CL models to three-factor and dual-factor TM, HM, CL models on the basis of the researches in chapter three at first. Domestic studies mostly only focus on single-factor TM and HM models. This paper utilizes nine models and examines different models'sensitiveness to single-factor and multifactor benchmarks. Empirical results indicate that the consistency of similar models (TM, HM or CL) is relatively stronger, but there are certain differences among the results of different kinds of models. The fitting ability of dual-factor model improves evidently than single-factor and three-factor models, and its explanation ability is stronger than three-factor model. The funds favor stocks with large scales, but they have no favor for growth stocks or value stocks. In general, Chinese funds don't possess the ability of selecting stocks and marketing time. Then this paper adopts the nonparametric test put forward by Jiang in 2003 to test the ability of market timing of Chinese funds. The results indicate that no managers of Chinese funds possess evident ability of market timing, which is different with the results of parametric methods in a sort. But in terms of Chinese funds on the whole, the results of both parametric and nonparametric methods indicate that Chinese funds have no significant ability of security selecting and market timing. Finally, theempirical results of Fame's fund performance decomposition model indicate that Chinese funds have certain ability of selecting stocks. But because Fama's fund performance decomposition depends on the validity of CAPM and Jensen model and ignores the significance test of independent variables, so we should treat the results with reservations. By contrast, the evaluation results of TM, HM and CL models are more robust. Synthesizing several methods, Chinese funds neither possess the remarkable ability of selecting stocks, nor the remarkable ability of marketing time. Chapter 5 studies the fund performance persistence. Minority scholars in China use cross-section regression, cross product ratio test and Spearman grade correlation coefficient test to carry on the empirical analysis of Chinese funds, but these studies mostly base on single period and single performance measure, which reduce the evaluation results'robustness and the conclusion'creditability. This paper introduces the test produced by Malkiel in 1995 for the first time. Concerning the methods on the basis of contingency table, besides defining winner and loses according to the compare of the fund performance and the median of all funds'performance, this paper redefines winner and loses by comparing the fund performance with zero, thus we studies the persistence of Chinese funds'relative and absolute performance. Then basing on the rolling sample periods, this paper selects five performance measurement including pure return, the fund performance relative to the market benchmark, Jensen Index based on single-factor, dual-factor and three-factor models and carries out the empirical analysis through cross-section regression, the methods on the basis of contingency table and Spearman grade correlation coefficient test. As regards Chinese funds'absolute performance persistence, the results of methods based on contingency table and cross-section regression are consistent. Chinese funds'performance isn't persistent during one month because winner and loser's performance aren't both persistent. The funds'performance has weak persistence in one quarter and half a year because winner'performance has some persistence and loser's performance is reversal. The funds'performance has no significant persistence in one year because winner'performance has significant persistence and loser's performance is significantly reversal. As regards Chinese funds'relative performance persistence, the results of methods based on contingency table and Spearman grade relation test are consistent. Both methods draw the conclusion that the funds'performance isn't significantly persistent in one month, one quarter, half a year and one year. In a word, the relative and absolute performance of Chinese funds on a whole has no significant persistence, which produces the evidence of the weak-form effectiveness of Chinese securities market.Chapter 6 studies the comprehensive quantitative evaluation of the fund performance. Only a few scholars have gone on researches at present in China, but most of their researches have shortcomings in some degree, such as only carrying on qualitative analysis, or irrational indexes chosen, or comprehensive methods being too subjective and simple. Considering the characteristics of Chinese fund market, this paper sets up overall and rational fund performance evaluation system, and puts forward three comprehensive quantitative methods for evaluating the fund performance. In this paper, the comprehensive evaluation system of fund performance consists of five aspects including the overall fund performance, manager's investment ability, fund expenses, fund liquidity and fund growth ability, which consists of sixteen indexes. On this basis, in order to drop multidimensional indexes to one-dimensional index, we utilize combined method consisting of analytical hierarchy process (AHP) and gray related analysis (GRA), data envelopment analysis (DEA), and principal components analysis to carry out the empirical studies on Chinese funds. Especially, combined method consisting of AHP and GRA is put forward for the first time to evaluate fund performance by the author, which is a new try for quantitatively evaluating the fund comprehensive performance. The empirical results show that three comprehensive methods have stronger consistency. The combined method consisting of AHP and GRA and PCA have very strong relationship, but the relationship between DEA and other methods are slightly lower, but still remarkable. In this paper, according to the request of DEA, there are some difference among the indexes chosen for DEA and the other two methods to some extent, which leads to the difference of DEA with the other two methods. If the indexes chose for the three methods are more unanimous, we believe the results of three methods are more similar. The empirical results indicate that the funds with outstanding comprehensive performance mainly include fund Kehui, Kexiang, Jinsheng, Yuze and Yifangda Stability, and the funds with worst comprehensive performance mainly include fund Puhua, Jingbo, Anrui, Hongyang and Tianhua. Then we simply averages ranks of every fund belonging to different fund management corporations, and gets the rank of different fund corporations, and thus carries out the comparative studies of different fund corporations. The empirical results of different methods have strong consistency. Yifangda, South and Boshi (especially Yifangda) fund management corporations have outstanding management level, but Penghua, Baoying and Dacheng fund corporations'management level are worst. The innovations of this paper mainly include the following aspects. Firstly, we carry on further analysis to the assumption and suitability of various methods, point out the mistakes caused by improperly using methods especially the classical methods. Secondly,...