The Study of Domestic Equity Fund Styles and Performance Persistence

• Main Authors: Tseng, Shao-Fang, 曾少芳
• Other Authors: Yang Chau-Chen
• Format: Others
• Language: zh-TW
• Published: 1997
• Online Access: http://ndltd.ncl.edu.tw/handle/98698168497698697934

碩士 === 國立臺灣大學 === 財務金融學系 === 85 === Due to the multiplicity of investment tools,there are more and more people using mutual funds as an investment tool nowadays. From the fund investors'' point of view, because of limited information, it isreally difficult for them to select a satisfying fund to invest froma variety of names and types. Although fund investors care about the profitability potential of funds most, we wonder whether the past performance evaluation can be a reference to invest. If the funds well-performed in the past, and thereafter they have at least as good performance as before, performance evaluation results can be an useful information to fund investors. However, the evaluation must be under the assumption that all funds be compared within the same style or similar strategies.The purpose of this study is to classify the funds into styles and examine the performance persistence. This study focused on 40 domestic closed-end and open-end equity funds for the period of 1995 and 1996.All of the data are monthly based. In order to classify the 40 fundsinto styles, this stydy weight-averaged the funds'' total stock holding,shares outstanding, market value, P/B ratio, and dividend yield, and then used these characteristic variables to perform two-stage clusteringmethod. Besides, this study applied the one-way ANOVA to test the average performance equality under each style and among the whole sample by Spearman rank correlation coefficient test.Under the span of the sample period, the conclusions of this study showed that:1. Sample funds were classified into 6 styles. They were aggressive growth, growth, growth and income, and trend-chaser. Under the aggressive growth and trend-chaser styles, they were further classified into two subgroups respectively.2. The difference between the pre-specified styles in prospectus and the classified styles in this study was significant.3. There did not exist different performance among styles.4. There did not exist performance persistence under each style.5. There did not exist performance persistence for the whole sample funds.