Predicting Excess Stock Returns Out of Sample: Can Anything Beat the Historical Average? - Cambell & Thompson Summary

Campbell & Thompson

Main findings:

* Find that predictive regressions beat the historical average return, once simple restrictions suggested by investment theory are imposed on the signs of coefficients and return forecasts. (Key of the paper is that restricted regressions perform considerably better than these unrestricted regressions – this is the fresh perspective offered by the researchers, not considered before. Predictive regressions without restrictions are worthless, as past research found).

* Contrary to Goyal and Welch (2007) saying historical average excess stock return forecasts future excess stock returns better than regressions of excess returns on predictor variables. (Essentially saying that you can do no better than a simple forecast based on the historical average stock return. Campbell and Thompson focus on disproving that claim).

* Other research argues that the poor out-of-sample performance of predictive regressions is a systemic problem, not confined to any one decade.

* The out-of-sample explanatory power is small, but nonetheless is economically meaningful for investors.

* Investors can profit using market timing strategies.

* Performance of the models does not depend sensitively on the particular valuation ratio that is used or the manner in which it is adjusted for long-run growth.

* Illustrating econometric principle that even false theoretical restrictions can be helpful in forecasting if they reduce the variance of a predictor more than they increase its bias.

Proposed suggestion for future development:

* Explore alternative methods to combine theory with historical data.

Limitations:

* Gains from market timing strategies could be offset by the additional costs implied by those strategies. Paper doesn’t explore the issue with transactions costs. (Thus perhaps in reality this profitable technique is ‘eaten’ by transaction costs).

Results:

* Shows R2 values, dependant on...