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, B. Frot, +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.

    Spectral Deconfounding via Perturbed Sparse Linear Models
    • 8
    • PDF
    The Blessings of Multiple Causes
    • 64
    • PDF
    Deconfounding and Causal Regularization for Stability and External Validity
    Rejoinder: Invariance, Causality and Robustness (2018 Neyman Lecture)

    References

    SHOWING 1-10 OF 31 REFERENCES
    High-dimensional graphs and variable selection with the Lasso
    • 2,859
    • PDF
    Covariance regularization by thresholding
    • 946
    • PDF
    Large Covariance Estimation by Thresholding Principal Orthogonal Complements.
    • 481
    • Highly Influential
    • PDF
    Sparse inverse covariance estimation with the graphical lasso.
    • 3,733
    • Highly Influential
    • PDF
    Optimal Shrinkage of Eigenvalues in the Spiked Covariance Model.
    • 103
    • PDF
    Asymptotic normality and optimalities in estimation of large Gaussian graphical models
    • 161
    • PDF
    Latent variable graphical model selection via convex optimization
    • 86
    • PDF
    High-Dimensional Statistics: A Non-Asymptotic Viewpoint
    • 464