We investigate return predictability and the implied intertemporal hedging demands for stocks and bonds in the U.S., Australia, Canada, France, Germany, Italy, and U.K. We first estimate predictive regression models for domestic bill, stock, and bond returns in each country, where returns depend...

We develop metrics based on Shapley values for interpreting time-series forecasting models, including "black-box" models from machine learning. Our metrics are model agnostic, so that they are applicable to any model (linear or nonlinear, parametric or nonparametric). Two of the metrics,...