Wu, Y and Hernández-Lobato, JM and Ghahramani, Z Dynamic Covariance Models for Multivariate Financial Time Series. (Unpublished)Full text not available from this repository.
The accurate prediction of time-changing covariances is an important problem in the modeling of multivariate financial data. However, some of the most popular models suffer from a) overfitting problems and multiple local optima, b) failure to capture shifts in market conditions and c) large computational costs. To address these problems we introduce a novel dynamic model for time-changing covariances. Over-fitting and local optima are avoided by following a Bayesian approach instead of computing point estimates. Changes in market conditions are captured by assuming a diffusion process in parameter values, and finally computationally efficient and scalable inference is performed using particle filters. Experiments with financial data show excellent performance of the proposed method with respect to current standard models.
|Uncontrolled Keywords:||stat.ME stat.ME stat.ML|
|Divisions:||Div F > Computational and Biological Learning|
|Depositing User:||Cron Job|
|Date Deposited:||09 Dec 2016 17:20|
|Last Modified:||23 Apr 2017 02:17|