Some New Variance Bounds for Asset Prices
14 Pages Posted: 21 Dec 2004 Last revised: 2 Jan 2023
Date Written: December 2004
Abstract
When equity prices are determined as the discounted sum of current and expected future dividends, Shiller (1981) and LeRoy and Porter (1981) derived a relationship between the variance of the price of equities, p(t), and the variance of the ex post realized discounted sum of current and future dividends: p*(t): Var(p*(t))>= Var(p(t)). The literature has long since recognized that this variance bound is valid only when dividends follow a stationary process. Others, notably West (1988), derive variance bounds that apply when dividends are nonstationary. West shows that the variance in innovations in p(t) must be less than the variance of innovations in a forecast of the discounted sum of current and future dividends constructed by the econometrician, p^(t). Here we derive a new variance bound when dividends are stationary or have a unit root, that sheds light on the discussion in the 1980s of the Shiller variance bound: Var(p(t)-p(t-1)) >= Var(p*(t)-p*(t-1))! We also derive a variance bound related to the West bound: Var(p^(t)-p^(t-1)) >= Var(p(t)-p(t-1)).
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
Tests of Equal Forecast Accuracy and Encompassing for Nested Models
-
Long Swings in the Exchange Rate: are They in the Data and Do Markets Know it?
-
Exchange Rates and Fundamentals
By Charles M. Engel and Kenneth D. West
-
Exchange Rates and Fundamentals
By Charles M. Engel and Kenneth D. West
-
Empirical Exchange Rate Models of the Nineties: Are Any Fit to Survive?
-
Exchange Rates and Monetary Fundamentals: What Do We Learn from Long-Horizon Regressions?
By Lutz Kilian
-
Empirical Exchange Rate Models of the Nineties: Are Any Fit to Survive?
By Yin-wong Cheung, Menzie David Chinn, ...