This paper studies the predictability of stock returns using monthly data on eight markets over the period 1876-1913. In contrast to much of the existing literature I find broad predictability across stock markets. Market interest rates and seasonal dummies generally have predictive power, and...

One of the most puzzling findings in asset pricing is that expected returns dominate variation in the dividend-to-price ratio, leaving little room for dividend growth rates. Even more puzzling is that this dominance only emerged after 1945. We develop a present value model to argue that a...

The loss function in supervised deep learning is a key element for training AI algorithms. For models aiming at predicting asset returns, not all prediction errors are equal in terms of impact on the efficiency of the algorithm. Indeed, some errors result in poor investment decisions while other...

We introduce a Nelson-Siegel type interest rate term structure model with the underlying yield factors following autoregressive processes revealing time-varying stochastic volatility. The factor volatilities capture risk inherent to the term structure and are associated with the time-varying...

This paper finds positive evidence of return predictability and investment gains for individual corporate bonds for an extended period from 1973 to 2017. Our sample consists of both public and private company bond observations. We have implemented multiple machine learning methods and designed a...

We test the out-of-sample predictive power for one-year bond excess returns for a variety of models that have been proposed in the literature. We find that these models perform well in sample, but have worse out-of-sample performance than the historical sample mean. We write the one-year excess...

A rapidly growing literature has documented important improvements in volatility measurement and forecasting performance through the use of realized volatilities constructed from high-frequency returns coupled with relatively simple reduced-form time series modeling procedures. Building on...

Testing procedures for predictive regressions with lagged autoregressive variables imply a suboptimal inference in presence of small violations of ideal assumptions. We propose a novel testing framework resistant to such violations, which is consistent with nearly integrated regressors and...