Predictive Regressions

50 Pages Posted: 17 Mar 2000 Last revised: 27 Feb 2023

See all articles by Robert F. Stambaugh

Robert F. Stambaugh

University of Pennsylvania - The Wharton School; National Bureau of Economic Research (NBER)

Date Written: May 1999

Abstract

When a rate of return is regressed on a lagged stochastic regressor, such as a dividend yield, the regression disturbance is correlated with the regressor's innovation. The OLS estimator's finite-sample properties, derived here, can depart substantially from the standard regression setting. Bayesian posterior distributions for the regression parameters are obtained under specifications that differ with respect to (i) prior beliefs about the autocorrelation of the regressor and (ii) whether the initial observation of the regressor is specified as fixed or stochastic. The posteriors differ across such specifications asset allocations in the presence of estimation risk exhibit sensitivity to those differences.

Suggested Citation

Stambaugh, Robert F., Predictive Regressions (May 1999). NBER Working Paper No. t0240, Available at SSRN: https://ssrn.com/abstract=205390

Robert F. Stambaugh (Contact Author)

University of Pennsylvania - The Wharton School ( email )

The Wharton School, Finance Department
University of Pennsylvania
Philadelphia, PA 19104-6367
United States
215-898-5734 (Phone)
215-898-6200 (Fax)

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States