Title: Expectation Puzzles, Time-varying Risk Premia, and Dynamic Models of the Term Structure
Abstract: TGDL#VWHUQQ\XHGX NHQ#IXWXUHVWDQIRUGHGX1 The mnemonic LPY stands for "sample-based Linear Projection coefficients in Yield-based regressions."Similarly, MPY stands for "Model-based LPY's."There are two versions of MPYs, one with risk-premium adjustment, and the other one without.2 We are presuming that the historical pattern LPY is not spurious, but rather is representative of the pattern of the population φ n .As demonstrated by Bekaert, Hodrick, and Marshall [1997a] and Backus, Foresi, Mozumdar, and Wu [1997], the patterns LPY cannot be attributed to small-sample bias in the relevant linear projections.Indeed, they find that the small-sample bias reinforces the puzzle by making the projection coefficients under LPY less negative than they would be in the absence of such bias.3 We are grateful to Backus, Foresi, Mozumdar, and Wu [1997] for providing the smoothed Fama-Bliss