Sparse Bayesian time-varying covariance estimation in many dimensions

By Gregor Kastner. Journal of Econometrics, 2016

105 citations recorded by the source corpus.

Dynamic covariance estimation for multivariate time series suffers from the curse of dimensionality. This renders parsimonious estimation methods essential for conducting reliable statistical inference. In this paper, the issue is addressed by modeling the underlying co-volatility dynamics of a time series vector through a lower dimensional collection of latent time-varying stochastic factors. Furthermore, we apply a Normal-Gamma prior to the elements of the factor loadings matrix. This hierarchical shrinkage prior effectively pulls the factor loadings of unimportant factors towards zero, thereby increasing parsimony even more. We apply the model to simulated data as well as daily log-returns of 300 SP it is implemented in the R package factorstochvol. (author's abstract)

Subfields: Computer Vision and Pattern Recognition, Finance, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/9a63fc7832d6c8bc7caf990d5960b7fd09d82b05

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Joshua C. C. Chan
2. Florian Huber
3. Martin Feldkircher

In Finance

1. Gary Koop
2. Xuewen Yu
3. Joshua C. C. Chan
4. Florian Huber
5. Martin Feldkircher
6. Mark F. J. Steel
7. Gregor Kastner
8. David Gabauer

In Statistics and Probability

1. Gary Koop
2. Joshua C. C. Chan
3. Florian Huber
4. Martin Feldkircher
5. Mark F. J. Steel
6. R. Horvath

Citations in the corpus

• Adaptive Shrinkage in Bayesian Vector Autoregressive Models (2019, methodology): “The first two modifications of the NG prior have recently been applied in Kastner (2016) in a factor stochastic volatility framework while the third modification is related to the multiplicative Gamma process prior proposed in Bhattacharya and Dunson (2011) within a static factor framework.”
• Large Order-Invariant Bayesian VARs with Stochastic Volatility (2021, background): “Kastner (2019) notes that, if one imposes the usual identification restrictions (e.g., factor loadings are triangular), then the model is not order invariant.”