Description Details Author(s) References
Conjugate Bayesian covariance matrix estimation using linear shrinkage with multiple target matrices (Gray et al., 2018). Most useful in high-dimensional data settings, where the number of variables is greater than the number of samples.
This package contains functions for covariance estimation using a conjugate Bayesian model. Whilst the main functionality of the package is for multiple target linear shrinkage estimation, we also provide functionality for the single target analogue (Hannart and Naveau, 2014; Gray et al., 2018).
These shrinkage methods perform best when an external dataset is used to create a target matrix/target matrices that is informative of the actual dataset under examination. An example of this utility is provided in Gray et al. (2018), in which high-dimensional protein covariance matrices for various cancer types are greatly informed by large sample covariance matrices from 'similar' cancer types.
Harry Gray
Maintainer: Harry Gray <h.w.gray@dundee.ac.uk>
Gray, H., Leday, G.G.R., Vallejos, C.A. and Richardson, S., 2018. Shrinkage estimation of large covariance matrices using multiple shrinkage targets. arXiv preprint.
Hannart, A. and Naveau, P., 2014. Estimating high dimensional covariance matrices: A new look at the Gaussian conjugate framework. Journal of Multivariate Analysis, 131, pp.149-162. doi.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.