Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Modelling of higher moments of stock return processes |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Economics, Business Administration and Information Technology |
| Date | 2008 |
| Abstract Text | Most popular models for Conditional Higher Moments of stock return processes are imposing a specific assumption on their parametric structure. This paper studies the issue of predicting Conditional Higher Moments of stock return processes and the associated Probability Densities with only making an assumption on the appropriate filtration. A Parameter Free Approach to Predictions for that purpose is introduced, that relies on Nonparametric Probability Estimations and the concept of Maximum Likelihood. The Prediction Approach is tested with artificial data from standard models for Conditional Volatility and Conditional Skewness as well as for observed market data. It is able to understand the properties of those standard models and is able to follow them, in case of the GARCH model unfortunately only with a slight bias. In case of observed market data the Approach is able to recognize some of the stylized facts, which are observed with Conditional Volatility. Furthermore it confirms the proposed property, that Conditional Skewness is reversely correlated with the lagged return. |
| Export |
BibTeX
|