Corpus ID: 84841712

# RSVP-graphs: Fast High-dimensional Covariance Matrix Estimation under Latent Confounding

@article{Shah2018RSVPgraphsFH,
title={RSVP-graphs: Fast High-dimensional Covariance Matrix Estimation under Latent Confounding},
author={Rajen Dinesh Shah and B. Frot and Gian-Andrea Thanei and N. Meinshausen},
journal={arXiv: Methodology},
year={2018}
}
• Rajen Dinesh Shah, +1 author N. Meinshausen
• Published 2018
• Mathematics
• arXiv: Methodology
• In this work we consider the problem of estimating a high-dimensional $p \times p$ covariance matrix $\Sigma$, given $n$ observations of confounded data with covariance $\Sigma + \Gamma \Gamma^T$, where $\Gamma$ is an unknown $p \times q$ matrix of latent factor loadings. We propose a simple and scalable estimator based on the projection on to the right singular vectors of the observed data matrix, which we call RSVP. Our theoretical analysis of this method reveals that in contrast to PCA-based… CONTINUE READING
5 Citations

#### Figures from this paper

The Blessings of Multiple Causes
• Computer Science, Psychology
• ArXiv
• 2018
• 70
• PDF

#### References

SHOWING 1-10 OF 31 REFERENCES
Large Covariance Estimation by Thresholding Principal Orthogonal Complements.
• Mathematics, Medicine
• Journal of the Royal Statistical Society. Series B, Statistical methodology
• 2013
• 504
• Highly Influential
• PDF
Sparse inverse covariance estimation with the graphical lasso.
• Mathematics, Medicine
• Biostatistics
• 2008
• 3,819
• Highly Influential
• PDF
Optimal Shrinkage of Eigenvalues in the Spiked Covariance Model.
• Mathematics, Medicine
• Annals of statistics
• 2018
• 105
• PDF