CAPM Over the Long-Run: 1926-2001
- Topics:
- Commercial Lending
FREE Registration is required
Overview: The CAPM can account for the spread in the average returns of portfolios sorted by book to market ratios over the long run from 1926-2001. In contrast, other studies document strong evidence of a book-to-market effect using post-1963 data, but they do so by relying on asymptotic standard errors. This article shows that making inferences that rely on asymptotic distributions is misleading. Using robust small sample inference, we fail to find a statistically significant book to market effect. The differences between the small sample and the asymptotic distributions can be attributed mostly to persistent time-varying betas. Rather than incorporating additional risk factors to explain these portfolio returns, the results suggest using a conditional CAPM with time-varying betas.
(Is this item miscategorized? Does it need more tags? Let us know.)
Format: PDF | Size: 366KB | Date: Sep 2003 | Pages: 46



