CHAPTER 14Bayesian α
Fool me once, shame on you. Fool me twice … well the point is we can’t get fooled again.
—George W. Bush
Whether from a qualitative research report or from knowledge passed along by word of mouth, portfolio managers usually have useful investing information that does not originate in a data set. Managers often attempt to build such information into an existing quantitative factor model by transforming it into the constant term α, but these attempts can be awkward. Assigning values to qualitative ideas can be somewhat arbitrary, and the new information may not combine efficiently with the estimation in the model. In such cases, Bayesian theory is extremely helpful. The theory, which provides a rigorous way ...