Speaker: Edward Qian.
Title: Cross-sectional Stock Selection Models.
Affiliation: Putnam Investments.
Date: Friday, February 22, 2002.
Place and Time: Room 499 - Dirac Science Library, 4:30 pm.

Abstract. One of the primary tools used by active equity portfolio managers and equity hedge fund managers is cross-sectional model. In contrast to time series model, the cross-sectional model does not aim to predict the return of individual stocks. Rather, it aims to predict the rank of relative returns of all stocks. Three factors determine the performance of a cross-sectional model: correlation between model forecasts and stock returns, dispersion of the forecasts and dispersion of the stock returns. We present a simple mathematical framework of cross-sectional model using the three factors and illustrate it with some practical examples.