Research On Stock Picking And Market Timing Abilities Of The Open-end Fund

In recent years,the public fund has showed explosive growth in the number and size in China,it has become an important investment product for investors.Because of the open-end stock and hybrid funds' flexible investment,good liquidity and high market transparency and other advantages,they are quite popular in the securities investment fund market.As the public funds are often managed by senior professionals,the amount of fund managers with the size of the fund has increased.The primary holder of our raised funds becomes institutional investors in China at the end of 2015.With the increase of fund size,types and management companies,how to evaluate the fund performance and the stock picking and market timing ability of fund managers effectively has attracted the attention of investors,regulators and researchers,and thus has important significance.The open-end stock and partial stock funds are mainly invested in the stock market,so the performance evaluation of the multi-factor pricing model based on the stock market anomalies has been paid attention by the industry and academia.With the study of scholars on the stock excess returns caused by different market pricing anomalies,the pricing factor selection is also constantly developed and rich.Carhart four-factor model as the first basic model,it expresses the fund performance comprehensively from four aspects: market risk factor,size factor,book-to-market ratio factor and momentum factor.It is widely used in China's stock market as a standard model,so it can provide strong theoretical and empirical basis for fund performance evaluation.The second basic model adopted in this paper is the four-factor pricing model(this is referred to as q-factor model),which includes market risk factor,size factor,investment factor and profit factor,is constructed by Hou,Xue and Zhang based to the investment q theory.The model has a certain empirical basis in developed markets,but in the emerging markets,especially in China's capital market,the empirical application is less.Therefore,the q-factor model is combined with the open-end fund market in China to evaluate the performance of the fund,which enriches the empirical research in the field of asset pricing in China.In this paper,the focus of the empirical analysis is to use the above two basic models to evaluate the sample funds,and analyze whether the conclusions are consistent.In addition,the q-factor model compared with the widely accepted Carhart four-factor model to explore the applicability of the model in China's open-end fund market.In recent years,scholars have been concerned about the benchmarks are also rising,and they believe standard four-factor models(including the standard Carhart four-factor model and the standard q-factor model)which ignore the performance of benchmarks has estimated for most funds incorrectly.Therefore,in this paper,the writer considers to eliminating the effect of the benchmark itself and uses the excess return which is based on the benchmark as an explanatory variable to constructed modified four-factor models(including the modified Carhart four-factor model and the modified q-factor model).This paper selects 437 open-end stocks and partial stock funds which were established before 2013 and did not contain passive index funds as objectives to be researched,and investigated the performance of the sample fund from 2005 to 2015.The fund is divided into groups and the corresponding style index is selected as the benchmark according to the style of investment.The modified four-factor models is used to analyze the fund performance.According to the alpha value to judges the fund's stock picking ability,based on the idea that the fund managers can change the sensitivity of fund portfolio to factors,this paper constructs the timing factor and analyzes the timing ability.And compared with the standard four-factor models to analyze the impact of the benchmark on the empirical results.This paper has the following findings through empirical research.(1)Most of the fund portfolios including the total sample have a negative accumulated monthly excess return on the basis of the benchmark,indicating that the fund's performance is worse than the benchmark at the average level.(2)Most of benchmarks(including comparative benchmarks and alternative benchmarks)show significant non-zero alpha values and factor coefficient values in the standard four-factor model,which indicates that benchmark have influence on the results of regression.(3)The alpha values of the total samples are not significant in both the standard and modified q-factor models,and are significantly negative in both the standard and modified Carhart four-factor models.The regression results based on two basic models show that the fund managers do not have stock picking ability at the average level.(4)Most of the fund portfolios have significant positive short-term timing factors and negative long-term timing factors in the modified q-factor model.And in the modified Carhart four-factor model,they show significant negative short-term timing factor and positive long-terming timing factors,but the majority of portfolios that including the total sample,the total timing factor based on the two basic models are not significant,that the fund managers in the average level do not have a significant total timing ability.