Structural Learning with Time-Varying Components: Tracking the Cross-Section of Financial Time Series
Authors:
- 24 May 2005
- journal article
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 67 (3) , 321-341
Abstract
Summary: When modelling multivariate financial data, the problem of structural learning is compounded by the fact that the covariance structure changes with time. Previous work has focused on modelling those changes by using multivariate stochastic volatility models. We present an alternative to these models that focuses instead on the latent graphical structure that is related to the precision matrix. We develop a graphical model for sequences of Gaussian random vectors when changes in the underlying graph occur at random times, and a new block of data is created with the addition or deletion of an edge. We show how a Bayesian hierarchical model incorporates both the uncertainty about that graph and the time variation thereof.Keywords
This publication has 16 references indexed in Scilit:
- Markov Chain Markov Field dynamics: Models and statisticsStatistics, 2002
- Monte Carlo Statistical Methods1999
- Positive Portfolio Factors1998
- Markov properties of nonrecursive causal modelsThe Annals of Statistics, 1996
- Hyper Markov Laws in the Statistical Analysis of Decomposable Graphical ModelsThe Annals of Statistics, 1993
- An approximation to maximum likelihood estimates in reduced modelsBiometrika, 1990
- Decomposition of maximum likelihood in mixed graphical interaction modelsBiometrika, 1989
- Algorithms for Minimum Coloring, Maximum Clique, Minimum Covering by Cliques, and Maximum Independent Set of a Chordal GraphSIAM Journal on Computing, 1972
- Covariance Selection1972
- On rigid circuit graphsAbhandlungen aus dem Mathematischen Seminar der Universitat Hamburg, 1961