Equity Premium Prediction Research Articles

Probability weighting and equity premium prediction: Investing with optimism

AbstractEmpirically motivated theoretical models of probability weighting which overweight tail events are finding many applications in finance. However, probability weighting has not yet been applied to equity premium prediction or to constructing optimal market timing investment strategies. We show that a measure of market optimism from a representative agent asset pricing model with probability weighting can be used to construct optimal dynamic investment strategies that outperform the buy‐and‐hold strategy and strategies generated by 17 leading equity premium predictors. We further show that this theory‐based measure of market optimism predicts the equity premium and market Sharpe ratio in‐sample and out‐of‐sample. The predictability is not subsumed by disaster probabilities, market sentiment, or market skewness. Our results indicate that our theory‐based measure provides a distinct channel for predicting aggregate stock returns.

Forecasting the equity premium: can machine learning beat the historical average?

We empirically predict the equity premium with the selected machine learning methods in Gu et al. (Empirical asset pricing via machine learning. Rev. Financ. Stud., 2020, 33(5), 2223–2273). We also consider four additional popular machine learning methods (ridge regression, support vector regression, k-nearest neighbors, and extreme gradient boosted trees) and their combination method. Using a dataset of both macroeconomic and technical predictors, we find that despite showcasing strong in-sample forecasting abilities, particularly with tree-based models, the out-of-sample results support Welch and Goyal (A comprehensive look at the empirical performance of equity premium prediction. Rev. Financ. Stud., 2008, 21(4), 1455–1508) that the competing forecasting models generally fail to outperform the historical average benchmark. We attribute this failure to the small dataset size and the low signal-to-noise ratio inherent in equity premium prediction. Our variable importance analysis further identifies three bond interest rate-related variables as the most dominant predictors for the equity premium. The economic value from a market timing perspective highlights that the historical average benchmark strategy generates the highest average return of 25.63% and the best Sharpe ratio of 0.8. Finally, our findings are robust across a variety of settings.

Predicting Equity Premium: A New Momentum Indicator Selection Strategy With Machine Learning

ABSTRACTWe propose a new momentum‐determined indicator‐switching (N‐MDIS) strategy, harnessing the power of machine learning to enhance the accuracy of equity premium prediction. Specifically, we re‐examine the regime‐dependent feature of univariate predictive regression relative to the benchmark. Furthermore, we investigate the prediction mechanism of the momentum‐determined indicator‐switching (MDIS) strategy and validate the significance of market regime information for the MDIS. Our findings demonstrate an overwhelmingly superior ex‐post forecasting performance compared with the MDIS. More notably, our empirical results substantiate that machine learning greatly aids in momentum indicator selection. The results show that the N‐MDIS with machine learning generates more accurate ex‐ante equity premium forecasts than both MDIS strategy and N‐MDIS strategy with logistic regression, yielding statistically and economically significant results. Moreover, our new approach exhibits robust forecasting performance across a series of robustness tests.

Out-of-sample equity premium predictability: An EMD-denoising based model

The poor out-of-sample forecasting performance of the stock returns of various predictors has been widely confirmed in the literature, which casts doubt on the reliability of stock-return predictability. However, the reliability of return predictability is closely related to the noise contained in the data. In this study, we design a new method to address the noise in the framework of empirical mode decomposition. The EMD method provides an efficient return decomposition, and based on which we selectively remove high-frequency components that are more likely to be contaminated by outliers. Our new model delivers statistically and economically significant out-of-sample gains relative to the historical average. The predictive ability mainly originates from the business-cycle risk and survives a series of robustness tests.

A Comprehensive 2022 Look at the Empirical Performance of Equity Premium Prediction

Abstract Our paper reexamines whether 29 variables from 26 papers published after Goyal and Welch 2008, as well as the original 17 variables, were useful in predicting the equity premium in-sample and out-of-sample as of the end of 2021. Our samples include the original periods in which these variables were identified, but end later. More than one-third of these new variables no longer have empirical significance even in-sample. Of those that do, half have poor out-of-sample performance. A small number of variables still perform reasonably well both in-sample and out-of-sample.

Bayesian Reconciliation of Return Predictability

Abstract This article considers a stable vector autoregressive (VAR) model and investigates return predictability in a Bayesian context. The bivariate VAR system comprises asset returns and a further prediction variable, such as the dividend-price ratio, and allows pinning down the question of return predictability to the value of one particular model parameter. We develop a new shrinkage type prior for this parameter and compare our Bayesian approach to ordinary least squares estimation and to the reduced-bias estimator proposed in Amihud and Hurvich (2004. “Predictive Regressions: A Reduced-Bias Estimation Method.” Journal of Financial and Quantitative Analysis 39: 813–41). A simulation study shows that the Bayesian approach dominates the reduced-bias estimator in terms of observed size (false positive) and power (false negative). We apply our methodology to a system comprising annual CRSP value-weighted returns running, respectively, from 1926 to 2004 and from 1953 to 2021, and the logarithmic dividend-price ratio. For the first sample, the Bayesian approach supports the hypothesis of no return predictability, while for the second data set weak evidence for predictability is observed. Then, instead of the dividend-price ratio, some prediction variables proposed in Welch and Goyal (2008. “A Comprehensive Look at the Empirical Performance of Equity Premium Prediction.” Review of Financial Studies 21: 1455–508) are used. Also with these prediction variables, only weak evidence for return predictability is supported by Bayesian testing. These results are corroborated with an out-of-sample forecasting analysis.

The Chinese equity premium predictability: Evidence from a long historical data

Our paper reexamines the predictability of macroeconomic variables in China over a long period. Contrary to the findings in developed markets that macroeconomic variables have poor in- and out-of-sample forecasting performance in predicting equity premiums, we find that five of nine macroeconomic variables have senior in- and out-of-sample predictability at monthly and longer horizons. Moreover, the forecasting results can generate significant economic value for investors. Our study provides new evidence for Chinese stock prediction using macroeconomic variables.

Out-of-sample equity premium prediction: The role of option-implied constraints

Out-of-sample equity premium prediction: The role of option-implied constraints

It Takes Two to Tango: Economic Theory and Model Uncertainty for Equity Premium Prediction

It Takes Two to Tango: Economic Theory and Model Uncertainty for Equity Premium Prediction

Equity premium prediction: The role of information from the options market

We examine the role of information from the options market in forecasting the equity premium. We provide evidence that the equity premium is predictable out-of-sample using a set of CBOE strategy benchmark indices as predictors. We use a range of econometric approaches to generate point, quantile, and density forecasts of the equity premium. We find that models based on option variables consistently outperform the historical average benchmark. In addition to statistical gains, using option predictors results in substantial economic benefits for a mean–variance investor, delivering up to a fivefold increase in certainty equivalent returns over the benchmark during the 1996–2021 sample period.

Equity premium prediction: Taking into account the role of long, even asymmetric, swings in stock market behavior

Through a novel approach, this paper shows that substantial change in stock market behavior has a statistically and economically significant impact on equity risk premium predictability both on in-sample and out-of-sample cases. In line with Auer’s “B ratio”, a “Bullish index” is introduced to measure the changes in stock market behavior, which we describe through a “fluctuation detrending moving average analysis” (FDMAA) for returns. We consider 28 indicators. We find that a “positive shock” of the Bullish Index is closely related to strong equity risk premium predictability for forecasts based on macroeconomic variables for up to six months. In contrast, a “negative shock” is associated with strong equity risk premium predictability with adequate forecasts for up to nine months when based on technical indicators.

Equity premium prediction using the price of crude oil: Uncovering the nonlinear predictive impact

Highly cited studies, such as Mork (1989), Hamilton (1996), Hamilton (2003) and Kilian and Vigfusson (2013) establish a nonlinear connection between movements in the real (nominal) price of crude oil and the real gross domestic product (GDP) growth rate. Based on the observation that crude oil price increases have a different impact on the real GDP growth rate than crude oil price decreases, the authors introduce various nonlinear transformations of the price of crude oil to account for the nonlinear predictive impact. Using predictive regressions à la Goyal and Welch (2008), a growing number of studies in recent years have conditioned on these nonlinear crude oil price measures to predict equity premium out-of-sample. By applying the dynamic rotation rule suggested in Zhu and Timmermann (2021), we demonstrate that the form of nonlinearity that helps improve relative out-of-sample point prediction accuracy the most does not have to do with crude oil price increases (decreases) relative to recent prices as perceived by the current literature, but with connecting the relative out-of-sample point prediction performance between the benchmark and the crude oil price-based model with movements in economic variables measuring equity market volatility, economic activity, uncertainty and financial conditions. The statistical evidence of out-of-sample equity premium predictability also translates to economic gains.

A New Test for Multiple Predictive Regression

Abstract We consider inference for predictive regressions with multiple predictors. Extant tests for predictability (especially for joint predictability) may perform unsatisfactorily and tend to discover spurious predictability as the number of predictors increases. We propose a battery of new instrumental variables-based tests which involve enforcement or partial enforcement of the null hypothesis in variance estimation. A test based on the few-predictors-at-a-time parsimonious system approach is recommended. Empirical Monte Carlos demonstrates the remarkable finite-sample performance regardless of numerosity of predictors and their persistence properties. Empirical application to equity premium predictability is provided.

Beating the Average: Equity Premium Variations, Uncertainty, and Liquidity

This paper contributes to the equity premium prediction literature by studying the performance of rarely or not researched predictors. To do so, we analyze the ability of state‐of‐the‐art liquidity and uncertainty predictors to beat the historical average when forecasting the monthly US equity premium. For this purpose, we apply an out‐of‐sample predictive regression approach to analyze statistical accuracy as well as economic gains of equity premium forecasts. Our findings show that the treasury‐eurodollar (TED) spread, as well as the macroeconomic uncertainty measure, is able to beat the historical average and provide robust predictions in various business cycles. Moreover, these two economic predictors also beat forecasts of a classical time series model.

Equity Premium Prediction: The Role of Informationfrom the Options Market

Equity Premium Prediction: The Role of Informationfrom the Options Market

Out-of-Sample Equity Premium Prediction: The Role of Option-Implied Constraints

Out-of-Sample Equity Premium Prediction: The Role of Option-Implied Constraints

Equity Premium Prediction Taking into Account the Role of Long, Even Asymmetric, Swings in Stock Market Behavior

Equity Premium Prediction Taking into Account the Role of Long, Even Asymmetric, Swings in Stock Market Behavior

Equity premium prediction: The role of information from the options market

Equity premium prediction: The role of information from the options market

Predicting equity premium with adjusted dividend-price ratio: the USA and international evidence

Purpose This paper aims to achieve two main objectives. The first is to introduce a suitable adjustment to the conventional dividend-price ratio, which would address econometric concerns and improve the predictability of the equity premium. The second is to compare the predictive performance of the newly introduced adjusted dividend-price ratio with the conventional dividend-price ratio. Design/methodology/approach The authors hypothesize that the adjusted dividend-price ratio will have better predictive power and forecasting quality for equity premium compared to the conventional dividend-price ratio. To test the hypothesis, the authors predict equity premium with both variables on a sample of 11 developed and emerging market indexes over a period spanning June 1995 to March 2017. To accommodate time variation in parameter values or structural breaks in the data, the authors conducted a fixed window rolling regressions using both variables. A variety of forecast techniques including magnitude and sign accuracy measures are applied to compare the performance of forecasts. Findings The adjusted dividend-price ratio is shown to be stationary and has both lower persistence and variability compared with the conventional dividend-price ratio. The authors find that the adjusted dividend-price ratio provides superior out-of-sample (OOS) performance compared to the conventional dividend-price ratio, for both size and sign accuracy, in forecasting equity premium for the majority of the countries in the sample. Research limitations/implications This paper introduces an easy-to-follow modification in the conventional dividend-price ratio that can be replicated by researchers and practitioners alike. However, the study has a limitation in that it does not capture the impact of dividend-paying firms within each index on the predictive ability of the adjusted dividend-price ratio. Practical implications The knowledge of equity premium predictability is important in implementing market-timing strategies and could be beneficial for portfolio and risk management. The newly introduced variable is easy to construct using widely available data without the need for complex econometric estimation. Investors can use this variable to predict equity premiums in international markets, both developed and emerging. The findings of this paper will be relevant to financial analysts, portfolio managers, investors and researchers in international finance. For example, by using the adjusted dividend-price ratio, investors would see up to 0.5% improvement in their OOS monthly forecasts of the equity premium. Originality/value To the best of the authors’ knowledge, this is the first paper that proposes adjustment in the conventional dividend-price ratio based on the past observations of the most recent quarter. In this way, the paper offers fresh insight that dividend-price ratio is still useful to predict equity premium albeit, after some adjustments and modifications. The findings of the paper would result in renewed interest in using the dividend-price ratio as a predictor of the equity premium.

Bayesian model averaging and the conditional volatility process: an application to predicting aggregate equity returns by conditioning on economic variables

This study revisits the topic of predicting aggregate equity returns out-of-sample by conditioning on economic variables through Bayesian model averaging (BMA). Besides simultaneously addressing parameter instability and model uncertainty, I suggest a new model feature, namely, predictors in a given model can also impact the dependent variable through the conditional volatility process. The suggested econometric framework is straightforward to implement without requiring simulation. Likewise, the user can easily decide, which aspects of the predictive channel should to be switched on, off or altered. I apply the suggested framework to the well-known [Goyal, A. and Welch, I., A comprehensive look at the empirical performance of equity premium prediction. Rev. Financial Stud., 2008, 21, 1455–1508] dataset. An extensive out-of-sample prediction evaluation demonstrates that averaging over predictor combinations in a model that allows lagged predictors to impact aggregate equity returns exclusively through the conditional volatility process results in statistically significant more accurate density predictions relative to the benchmark, especially when predicting the left tail of the conditional distribution. One also observes economic gains in favor of certain BMAs. Here, the BMA that allows predictors to impact equity returns through the conditional mean as well as the conditional volatility process is the top performer.