Predicting Excess Stock Returns Out of Sample: Can Anything Beat the Historical Average?

By John Y. Campbell, S. B. Thompson. 2008

2,578 citations recorded by the source corpus.

Goyal and Welch (2007) argue that the historical average excess stock return forecasts future excess stock returns better than regressions of excess returns on predictor variables. In this article, we show that many predictive regressions beat the historical average return, once weak restrictions are imposed on the signs of coefficients and return forecasts. The out-of-sample explanatory power is small, but nonetheless is economically meaningful for mean-variance investors. Even better results can be obtained by imposing the restrictions of steady-state valuation models, thereby removing the need to estimate the average from a short sample of volatile stock returns.

Subfields: Finance, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Finance

In Management Science and Operations Research

Citations in the corpus

Deep Learning for Predicting Asset Returns (2018, methodology): “Even if there is evidence that R2 OS is statistically significant, R 2 OS values are typically small for prediction models, but as Campbell and Thompson (2007) and Rapach et al.”
Sufficient Forecasting Using Factor Models (2015, methodology): “Akin to the in-sample R 2 typically used in linear regression, the out-of-sample R 2 , suggested by Campbell and Thompson (2008), is deﬁned here as where in each replication we use the second half of the time series as the testing sample.”
Volatility‐Managed Portfolios (2017, methodology): “We use the estimate from Campbell and Thompson (2008) who obtain a number around 0.”
Expectations of Returns and Expected Returns (2014, neutral): “16 See Ferson, Sarkissian, and Simin (2003), Stambaugh (1999), Welch and Goyal (2008), Campbell and Yogo (2006), and Campbell and Thompson (2008) for a discussion of the time-series properties and performance of stock market return predictors.”
Forecasting the direction of the US stock market with dynamic binary probit models (2011, methodology): “…however, how much the in-sample evidence should be emphasized in assessing overall return predictability because it does not guarantee out-of-sample predictability, as emphasized in many previous studies (see discussion, for example, in Goyal and Welch, 2004, and Campbell and Thompson, 2007).”
Forecasting the Equity Risk Premium: The Role of Technical Indicators (2014, positive): “Following convention, we standardize the individual predictors before computing the principal components.”
Equity Risk Premiums (ERP): Determinants, Estimation and Implications – The 2012 Edition (2011, background): “Campbell and Thompson (2008) disagree, noting that putting simple restrictions on the predictive regressions improve out of sample performance for many predictive variables.111 To answer this question, we looked at the implied equity risk premiums from 1960 to 2011 and considered four predictors of…”
Multifactor models and their consistency with the ICAPM (2012, neutral): “gTERM , gDEF , gDY , gRF , gPE , gVS , gCP represent the risk prices associated with the term-structure spread, default spread, market dividend yield, one-month Treasury bill rate, market price–earnings ratio, value spread, and the Cochrane-Piazzesi factor, respectively. The models are the baseline ”
Are Institutions Informed About News (2015, methodology): “Applying the Campbell and Thompson (2008) framework implies that observing institutional order flow would enable investors in individual stocks to proportionally increase their expected returns by more than 40%.”
Predictability of Currency Carry Trades and Asset Pricing Implications (2013, neutral): “Reported are (i) out-of-sample R's (Campbell and Thompson, 2008) and (ii) one-sided p-values for the MSPE-adjusted statistic (Clark and West, 2007), obtained with two or three predictors and in combination forecasts (Stock and Watson, 2004).”