Abstract

We develop a normative theory for constructing mean-variance portfolios robust to model misspecification. We identify two inefficient portfolios---an alpha'' portfolio, representing latent asset demand, that depends only on pricing errors and a beta'' portfolio that depends on factor risk premia, which when combined give mean-variance efficient portfolios. Our key insight is to recognize that the alpha and beta portfolios have different economic properties and therefore we treat misspecification in these portfolios using different methods. We demonstrate that our theoretical insights lead to an economically substantial and statistically significant improvement in out-of-sample portfolio performance, which increases with the number of assets.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call