Inference
in Infinite Superpositions of Non-Gaussian Ornstein-Uhlenbeck Processes
Using Bayesian Nonparametric Methods
Jim Griffin (University of Kent)
18/03/11
This talk will describe a Bayesian nonparametric approach to volatility
estimation in financial time series. Volatility is assumed to follow a
superposition of an infinite number of Ornstein-Uhlenbeck processes
driven by a compound Poisson process with a nonparametric jump size
distribution. This model allows a wide range of possible temporal
dependencies and marginal distributions for volatility. The properties
of the model and prior specification for Bayesian inference will be
discussed. The model is fitted to daily returns of several indices.
Paper is available from here
Back to the seminar
homepage
LSE Home Page |
Departmental
Home Page
[Last modified: Mar. 15th 2011 by Kostas
Kalogeropoulos]