Predicting bond betas using macro-finance variables

We conduct in-sample and out-of-sample forecasting using the new approach of combining explanatory variables through complete subset regressions (CSR). We predict bond CAPM betas and bond returns conditioning on various macro-finance variables. We explore differences across long-term government bonds, investment grade corporate bonds, and high-yield corporate bonds. The CSR method performs well in predicting bond betas, especially in-sample, and, mainly high-yield bond betas when the focus is out-of-sample. Bond returns are less predictable than bond betas.