The Pricing Kernel Anomaly: the Case of the Information that did not Bark
Event details
| Date | 22.10.2015 |
| Hour | 12:00 › 13:00 |
| Speaker | Carlo SALA (PhD student, Università della Svizzera italiana) |
| Category | Conferences - Seminars |
This paper provides an empirical study on the conditioning of the information in the estimation of the pricing kernel. Although the neoclassical theory requires the pricing kernel to be a monotonically decreasing function, recent empirical studies found several violations: at the extremes as well as in the central part of the functional. Since Jackwerth (2000), this is known as the "pricing kernel puzzle". While the risk-neutral moments extracted from option surfaces are by construction forward looking, the ones obtainable from historical returns are only partially informative, thus suboptimal with respect to investors' future beliefs. This underestimation of the physical ltration produces a disalignment with respect to the full conditioning of the information set as required by the neoclassical theory. Empirically it turns out that most of papers present in literature are then aected by a non-homogeneity bias. We propose a new exible and highly informative non-parametric method to estimate a non-stationary and fully-conditional physical measure. Exploiting the informational content of the implied moments of option prices, the proposed measure embeds the missing forward looking information necessary to produce a time-varying and fully-conditional physical measure. A natural approach to exploit simultaneously multiple data and provide statistical inference is the Dirichlet process. Using the precision parameter of the Dirichlet process as a proxy for the missing information, we calibrate it with respect to the daily liquidity of the options in the market. The obtained density is a mixture of the two measures and combines the prior forward looking information available from option data with the historical background provided by stock returns. The new measure is then used to investigate the pricing kernel monotonicity.
Practical information
- Informed public
- Invitation required