How to measure risk in asset pricing models: entropy or beta?
| Authors | |
|---|---|
| Year of publication | 2017 |
| Type | Article in Proceedings |
| Conference | Enterprise and Competitive Environment Conference Proceedings |
| MU Faculty or unit | |
| Citation | |
| Web | https://ece.pefka.mendelu.cz/sites/default/files/imce/ECE2017_fin.pdf |
| Field | Management and administrative |
| Keywords | entropy; risk measure; beta; asset pricing |
| Description | Financial theory borrows scientific methods from natural sciences. In this paper, we consider one of such methods called entropy, which in financial terms can be considered as a measure of risk in asset pricing models. We propose three different non-parametric estimation techniques to estimate financial entropy, the results of which we compare to the CAPM beta based on their explanatory power to describe the diversity in expected risk premiums. Kernel density estimated Shannon entropy provides the most efficient results not dependent on the choice of the market benchmark and without imposing any prior model restrictions. Kernel density estimated Rényi entropy and maximum likelihood estimated Shannon entropy also perform better in-sample than the CAPM beta. |
| Related projects: |