Research on Predicting Equity Returns with Valuation Measures
Summary
The document asks whether combining multiple valuation measures, such as earnings multiples and dividend-discount estimates, can help time broad equity markets or forecast stock returns relative to fixed income. It also raises the possibility of combining valuation with momentum as part of tactical asset allocation. The response points readers to research on out-of-sample excess stock return prediction, including work responding to claims that such returns cannot be forecast reliably, and studies reviewing stock-return predictability.
A theoretical paper is also recommended for connecting return predictability to candidate forecasting variables. The answer frames these works as a starting point for understanding variables, tests, and theory, then suggests following later citations to locate more recent research. It does not summarize empirical findings, compare valuation models, or establish that any particular combination improves timing. Accordingly, the material is a short research roadmap rather than evidence for a deployable market-timing strategy; conclusions require reading the cited studies and evaluating their methods and out-of-sample performance.
Key ideas
- The question concerns using multiple valuation measures to forecast broad equity returns or returns relative to bonds.
- The response recommends foundational research on out-of-sample excess return prediction and stock-return predictability.
- A theoretical framework can help relate forecasting variables to the economic basis for return predictability.
- The document gives references to investigate, but reports no comparative results or validated valuation-and-momentum strategy.
Tags
Full text
# Which valuation measures are most useful for equity market timing? # Which valuation measures are most useful for equity market timing? Competing academic studies, such as Asness's Fight the Fed Model and Lee, Myers, and Swaminathan's What is the Intrinsic Value of the Dow, offer differing answers to the question of whether equity valuation measures (such as P/E in the case of Asness, DDM in the case of Lee et. al.) can be used to predict the direction of overall equity markets. These are just two seminal studies in the field of market timing, which is itself a part of the broader Tactical Asset Allocation literature. I would like to know if there are any academic studies (the more recent the better) which use more than one valuation measure/model to try to predict equity market returns, either in absolute terms or relative to fixed income. It would also be interesting if these studies look into what kinds of models may be combined with valuation, such as momentum (see Faber's A Quantitative Approach to Tactical Asset Allocation), in order to yield the best results. ## Answer by Ram Ahluwalia (score 4, accepted) https://quant.stackexchange.com/a/1703 Take a look at Campbell's 2008 paper "Predicting Excess Stock Returns out of Sample". This paper is in response to Goyal & Welch's paper which argued that excess returns cannot be predicted out of sample. Also see Baekart and Ang's paper "Stock Return Predictability: Is it there?". A good theoretical framework that ties stock return predictability to variables most likely to predict returns is Cochrane's 2008 paper "The Dog that did not bark: A defense of return predictability". You can identify more recent citations of these papers on Google Scholar for the latest research. However, I would suggest starting here since the key variables, tests, and theory are laid out by the more highly regarded academics.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.