Estimating the inverse of the covariance matrix, the precision matrices.

Two big data problems problems can arise here: large \(p\) (ambient dimension) and large \(n\) (sample size). Large \(p\) is a problem because the covariance matrix is a \(p \times p\) matrixand frequently we need to invert it to calculate some target estimand.

## The obvious way

Estimate the covariance matrix then invert it. This is the baseline. π

## QUIC

## Bayesian

π Wishart priors?

## Penalized

## Structured

## Bound distance from Covariance

covariance matrix or

## References

Aragam, Bryon, Jiaying Gu, and Qing Zhou. 2017. βLearning Large-Scale Bayesian Networks with the Sparsebn Package.β

*arXiv:1703.04025 [Cs, Stat]*, March.Avagyan, Vahe, and Xiaoling Mei. 2022. βPrecision Matrix Estimation Under Data Contamination with an Application to Minimum Variance Portfolio Selection.β

*Communications in Statistics - Simulation and Computation*51 (4): 1381β1400.Chen, Xiaohui, Mengyu Xu, and Wei Biao Wu. 2013. βCovariance and Precision Matrix Estimation for High-Dimensional Time Series.β

*The Annals of Statistics*41 (6).Fan, Jianqing, Yuan Liao, and Han Liu. 2016. βAn Overview of the Estimation of Large Covariance and Precision Matrices.β

*The Econometrics Journal*19 (1): C1β32.Hsieh, Cho-Jui, MΓ‘tyΓ‘s A. Sustik, Inderjit S. Dhillon, and Pradeep D. Ravikumar. 2014. βQUIC: Quadratic Approximation for Sparse Inverse Covariance Estimation.β

*Journal of Machine Learning Research*15 (1): 2911β47.Hsieh, Cho-Jui, MΓ‘tyΓ‘s A. Sustik, Inderjit S. Dhillon, Pradeep Ravikumar, and Russell A. Poldrack. 2013. βBIG & QUIC: Sparse Inverse Covariance Estimation for a Million Variables.β In

*Advances in Neural Information Processing Systems*, 16. NIPSβ13. Red Hook, NY, USA: Curran Associates Inc.JankovΓ‘, Jana, and Sara van de Geer. 2015. βHonest Confidence Regions and Optimality in High-Dimensional Precision Matrix Estimation.β

*arXiv:1507.02061 [Math, Stat]*, July.Khoshgnauz, Ehsan. 2012. βLearning Markov Network Structure Using Brownian Distance Covariance.β

*arXiv:1206.6361 [Cs, Stat]*, June.Kuismin, Markku O., and Mikko J. SillanpΓ€Γ€. 2017. βEstimation of Covariance and Precision Matrix, Network Structure, and a View Toward Systems Biology.β

*WIREs Computational Statistics*9 (6): e1415.Lam, Clifford, and Jianqing Fan. 2009. βSparsistency and Rates of Convergence in Large Covariance Matrix Estimation.β

*Annals of Statistics*37 (6B): 4254β78.Le, Thien-Minh, and Ping-Shou Zhong. 2021. βHigh-Dimensional Precision Matrix Estimation with a Known Graphical Structure.β arXiv.

Meier, Alexander, Claudia Kirch, and Renate Meyer. 2020. βBayesian Nonparametric Analysis of Multivariate Time Series: A Matrix Gamma Process Approach.β

*Journal of Multivariate Analysis*175 (January): 104560.Mercer, A. McD. 2000. βBounds for AβG, AβH, GβH, and a Family of Inequalities of Ky Fanβs Type, Using a General Method.β

*Journal of Mathematical Analysis and Applications*243 (1): 163β73.Moscone, Francesco, Elisa Tosetti, and Veronica Vinciotti. 2017. βSparse Estimation of Huge Networks with a Block-Wise Structure.β

*The Econometrics Journal*20 (3): S61β85.Pleiss, Geoff, Jacob R. Gardner, Kilian Q. Weinberger, and Andrew Gordon Wilson. 2018. βConstant-Time Predictive Distributions for Gaussian Processes.β In. arXiv.

Pourahmadi, Mohsen. 2011. βCovariance Estimation: The GLM and Regularization Perspectives.β

*Statistical Science*26 (3): 369β87.Saad, Yousef. 2003.

*Iterative Methods for Sparse Linear Systems: Second Edition*. 2nd ed. SIAM.Sharma, Rajesh. 2008. βSome More Inequalities for Arithmetic Mean, Harmonic Mean and Variance.β

*Journal of Mathematical Inequalities*, no. 1: 109β14.Ubaru, Shashanka, Jie Chen, and Yousef Saad. 2017. βFast Estimation of \(tr(f(A))\) via Stochastic Lanczos Quadrature.β

*SIAM Journal on Matrix Analysis and Applications*38 (4): 1075β99.Wang, Lingxiao, Xiang Ren, and Quanquan Gu. n.d. βPrecision Matrix Estimation in High Dimensional Gaussian Graphical Models with Faster Rates,β 9.

Wu, Wei Biao, and Mohsen Pourahmadi. 2003. βNonparametric Estimation of Large Covariance Matrices of Longitudinal Data.β

*Biometrika*90 (4): 831β44.Yuan, Ming. 2010. βHigh Dimensional Inverse Covariance Matrix Estimation via Linear Programming.β

*The Journal of Machine Learning Research*11: 26.Zhang, T., and H. Zou. 2014. βSparse Precision Matrix Estimation via Lasso Penalized D-Trace Loss.β

*Biometrika*101 (1): 103β20.
## No comments yet. Why not leave one?