Sequential Learning, Predictability, and Optimal Portfolio Returns

Abstract

This paper finds statistically and economically significant out-of-sample portfolio benefits for an investor who uses models of return predictability when forming optimal portfolios. The key is that investors must incorporate an ensemble of important features into their optimal portfolio problem including time-varying volatility and time-varying expected returns driven by improved predictors such as measures of yield that include shares repurchase and issuance in addition to cash payouts. In addition, investors need to account for estimation risk when forming optimal portfolios. Prior research document a lack of benefits to return predictability, and our results suggest this was largely due to omitting time-varying volatility and estimation risk. We also study the learning problem of investors, documenting the sequential process of learning about parameters, state variables, and models as new data arrives.

Stock Return Prediction with Fully Flexible Models and Coefficients

“…The literature agrees that ignoring model uncertainty and instability impairs predictability (Stambaugh, 1999;Wachter and Warusawitharana, 2009;Dangl and Halling, 2012;Billio et al, 2013;Johannes et al, 2014;Wachter and Warusawitharana, 2015). Time-varying parameters in predictive regressions have been previously considered in the literature.…”

“…Evidence for time-varying volatility is strong as it generates fat-tailed return distribution for stock market (Johannes et al, 2014). Moreover, numerous studies find the important role timevarying volatility plays in predicting excess stock return (Johannes et al, 2014;Joscha and Schüssler, 2014).…”

“…Evidence for time-varying volatility is strong as it generates fat-tailed return distribution for stock market (Johannes et al, 2014). Moreover, numerous studies find the important role timevarying volatility plays in predicting excess stock return (Johannes et al, 2014;Joscha and Schüssler, 2014). As a consequence, following Koop and Korobilis (2012), we use exponentially weighted Moving Average (EWMA) to estimate H t , which updates at each time and can be approximated by a recursive form:…”

“…This is known as model uncertainty. Stock return determinants and hence forecasting models may also change at each time period, as demonstrated in Avramov (2002), Cremers (2002), Rapach et al (2010), Dangl and Halling (2012) and Johannes et al (2014). Specifically, Johannes et al (2014) argue that incorporating different model features, such as time-evolving expected returns, time-varying volatility and taking account of estimation error, is beneficial for out-of-sample gains.…”

Provas Académicas na FEUC

“…They considered combinations of univariate and multivariate models, in a Bayesian framework, and showed that their model outperforms the forecasts based on the historical mean equity premium, over a wide range of time periods. (Johannes et al, 2014) proposed a model to estimate the relation between the net payout ratio and the equity premium, in the US, that included both time-varying coefficients and time-varying volatility. They concluded that their model provides statistically and economically significant out-of sample portfolio benefits for a power utility investor.…”