직접-역회귀분석을 이용한 국내 주식형펀드의 장기성과분석

When CAPM-based empirical models are used to measure the abnormal returns, the presence of either benchmark error and/or nonsynchronous trading will generally cause the estimated beta to be biased downward, thus causing the estimated alpha to be biased upward. In order to test the presence of biases in the CAPM-based abnormal return evaluation, this study introduces the direct and reverse regression methodology originally developed by Frich(1934). In the presence of these types of measurement errors, the direct and reverse estimates will provide a lower bound and an upper bound in the unknown true beta, thus placing bound on the true alpha(the measure of abnormal return). Therefore measuring the abnormal returns based on a more conservative estimate(the lower bound of alpha) will add confidence in judging the presence of the abnormal returns, avoiding the misjudgment.The direct and reverse regression are applied to examine the performances of equity-type funds. The results show that equity-type funds with durations over three years have shown no significant abnormal returns. Also we have concluded that the past performances do not effect on the performances of the following periods. Therefore this study concludes that the equity-type funds similarly follow the movement of the market and their performances have no correlation with the ability of the fund managers.