Abstract
dc:description.abstract<p>Due to the advancements in computing power and the availability of high-frequency data, the analyses of the high frequency stock data and market microstructure has become more and more important in econometrics. In the high frequency data setting, volatility is a very important indicator on the movement of stock prices and measure of risk. It is a key input in pricing of assets, portfolio reallocation, and risk management. In this thesis, we use the Heterogeneous Autoregressive model of realized volatility, combined with Bayesian inference as well as Markov chain Monte Carlo method’s to estimate the innovation density of the daily realized volatility. A Dirichlet process is used as the prior in a countably infinite mixture model. The semi-parametric model provides a robust alternative to the models used in the literature. I find evidence of thick tails in the density of innovations to log-realized volatility.</p>
Degree
thesis:*- Department dc:contributor.department
- Statistics
- Year dc:date.issued
- 2013
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Feng, Tian
- Advisors dc:contributor.advisor
-
- Maheu, John
- Childs, Aaron
Subjects
dc:subject × 6Identifiers
dc:identifier.*- Identifier URI
- https://doi.org/10.71548/11999
- Dc Identifier Other
-
opendissertations/8348
9360
4615928