Estimation and Inference in Low Frequency Factor Model Regressions with Overlapping Observations

A low frequency factor model regression uses returns computed at a lower frequency than data available. An example is using monthly rather than daily returns to estimate the Capital Asset Pricing Model (CAPM). I show that when using overlapping observations to estimate low frequency factor model regressions, the resulting slope coefficients have variances that are smaller by a factor of approximately (2/3), relative to slope coefficients estimated using non-overlapping observations. Furthermore, using overlapping observations results in improved estimates of the standard errors of both the intercept and slope coefficients. I present a simulation study and empirical examples of the CAPM and Fama French three factor model to support the theoretical results