As the wealth of societies increase, the demand for more specialized financial services grows. A good example is the emergence of professionally managed investment funds, PMFs, that seek to apply modern financial investment practices that enable individual investors to better manage their wealth. PMFs offer the benefits of diversification to investors by collecting small amounts of capital from a large number of investors and form diversified portfolios that are professionally managed. Investors benefit from a better risk-return ratio than they are likely to earn if they invest individually—at the cost of paying some managerial fees. In order to promote a more stable stock market, the Chinese Government intends to encourage more institutional investors to invest in the stock market. This has resulted in very significant growth in PMTFs over the past 5 years. The PMFs in China are mutual funds that are mainly closed-end funds (CEFs). There were 51 CEFs in June, 2002 with total capital under management of around 75 billion RMB. Except for three index funds, Xinghe, Pufeng and Jingfu, all of the others are managed funds that pursue growth stocks as their objective. Well known performance measures based on the Capital Asset Pricing Model, CAPM, such as the Sharpe index, Treynor index and Jensen index, are used in this study to examine the performance of CEFs in China. Several difficulties need to be overcome when using CAPM based measures. First, it’s hard to get the real market portfolio, and have to use an index as a proxy. According to the research of Lehmann and Modest (1987), the performance of mutual funds based on different benchmark portfolios is different, and sometimes the results are contradictory. Second, CAPM implies that beta is a constant coefficient, in fact, Admati & Ross (1985), Dybvig & Ross (1985) find the beta will change according to the manager’s different expectations about the market trend. Third, CAPM analysis is based on the assumptions that returns are normally distributed and investor preferences are adequately described by the mean and variance of the expected returns. In fact, the stock distributions can be abnormal and investors may require different returns for different market periods. The use of variance as a risk measure is also under challenge: variance can only measure the fluctuation of stock, not the direction of the fluctuation. However the right hand movement represents extra-profit to the investor—not risk. Despite these limitations, the use of CAPM based ratios is still worthwhile, however, new models will also be employed to overcome some of the problems. Following Roy’s Safety-first theory, left partial moment (LPM) is used to measure loss risk instead of variance. One aim of this paper is to find out if there is a most effective model for Chinese CEF. PMFs are a knowledge intensive industry. The role of a fund manager’s ability is a very important topic that is worth arguing. Thus, another aim for this research is to use historical data to measure whether Chinese fund managers have stock-selection and market-timing abilities.