Estimating the Dynamics of Mutual Fund

This article develops a Kalman filter model to track dynamic mutual fund factor loadings. It then uses the estimates to analyze whether managers with market-timing ability can be identified ex ante. The primary findings are as follows: (i) Ordinary least squares (OLS) timing models produce false positives (nonzero alphas) at too high a rate with either daily or monthly data. In contrast, the Kalman filter model produces them at approximately the correct rate with monthly data; (ii) In monthly data, though the OLS models fail to detect any timing among fund managers, the Kalman filter does; (iii) The alpha and beta forecasts from theKalmanmodel aremore accurate than those from the OLS timing models; (iv) The Kalman filter model tracks most fund alphas and betas better than OLS models that employ macroeconomic variables in addition to fund returns. (JELG12, G14, G23, C01, C12, C13, C52, C53) A great deal of attention has gone into understanding mutual fund returns and trading strategies.1 What most studies have in common is the maintained hypothesis that their returns can be represented by a static ordinary least squares (OLS) model.2 Although this may be a reasonabl