Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper presents evidence linking in-sample tests of predictive content and out-of-sample forecast accuracy. Our approach focuses on the negative effect that finite-sample estimation error has on forecast accuracy despite the presence of significant population-level predictive content. We derive in-sample tests that assess whether a variable has predictive content and whether this content is estimated precisely enough to improve forecast accuracy. Our tests are asymptotically non-central chi-square or non-central normal. We provide a convenient bootstrap for computing critical values. In Monte Carlo and empirical analysis, we examine the effectiveness of our testing procedure.