Asset pricing models: implications for expected returns and portfolio selection |
| |
Authors: | Craig MacKinlay A; Pstor L |
| |
Institution: | 1 The Wharton School, 3620 Locust Walk, Philadelphia, PA 19104-6367, USA
2 University of Chicago, Chicago, IL, USA
z Corresponding author
E-mail: acmack@wharton.upenn.edu |
| |
Abstract: | When a risk factor is missing from an asset pricing model, theresulting mispricing is embedded within the residual covariancematrix. Exploiting this phenomenon leads to expected returnestimates that are more stable and precise than estimates deliveredby standard methods. Portfolio selection can also be improved.At an extreme, optimal portfolio weights are proportional toexpected returns when no factors are observable. We find thatsuch portfolios perform well in simulations and in out-of-samplecomparisons. |
| |
Keywords: | |
本文献已被 Oxford 等数据库收录! |
|