Multi-target linear shrinkage estimation of large precision matrix

Yuang Xue, Lihong Wang

Article ID: 9912
Vol 7, Issue 2, 2024

VIEWS - 670 (Abstract)

Abstract


In this paper we propose a multi-target linear shrinkage estimator of the precision matrix by shrinking the inverse of the sample covariance matrix directly, which is a generalization of the single-target linear shrinkage estimator. The explicit expression of the weights of multi-target linear shrinkage estimator is derived when the ratio of the dimension p and the sample size n tends to a positive constant c ∈ (0, 1). The numerical simulation and an empirical analysis of financial market data are provided to compare the multi-target linear shrinkage estimator with some other estimators of the precision matrix proposed in the literature. The computation results show the improvement of the multi-target linear shrinkage estimator.

Keywords


frobenius norm; linear shrinkage estimator; multiple target matrices; precision matrix

Full Text:

PDF


References


Elton EJ, Gruber MJ, Brown SJ, Goetzmann WN. Modern Portfolio Theory and Investment Analysis. Wiley Publishing; 2009. Markowitz H. Portfolio selection. J. Finance. 1952; 7: 77–91. Bodnar O, Bodnar T, Parolya N. Recent advances in shrinkage-based high-dimensional inference. J. Multivar. Ana. 2022; 188: 104826. Bodnar T, Okhrin Y, Parolya N. Optimal shrinkage based portfolio selection in high dimensions. J. Bus. Econ. Stat. 2023; 41(1): 140–156. Fan J, Liao Y, Liu H. An overview of the estimation of large covariance and precision matrices. Econom. J. 2016; 19: C1–C32. Fan J, Liao Y, Mincheva M. High-dimensional covariance matrix estimation in approximate factor models. Ann. Stat. 2017; 39: 3320–3356. Friedman JH, Hastie T, Tibshirani R. Sparse inverse covariance estimation with the graphical lasso. Biostatistics. 2008; 9: 432–441. Ikeda Y, Kubokawa T, Srivastava MS. Comparison of linear shrinkage estimators of a large covariance matrix in normal and non-normal distributions. Comput. Statist. Data Anal. 2016; 95: 95–108. Kuismin MO, Kemppainen JT, Sillanpää MJ. Precision matrix estimation with ROPE. J. Comput. Graph. Statist. 2017; 26: 682–694. Ledoit O, Wolf M. Nonlinear shrinkage of the covariance matrix for portfolio selection: Markowitz meets goldilocks. Rev. Financ. Stud. 2017; 30: 4349–4388. Ledoit O, Wolf M. Shrinkage estimation of large covariance matrices: Keep it simple, statistician? J. Multivar. Anal. 2021; 186: 104796. Liu C, Tang CY. A quasi-maximum likelihood approach for integrated covariance matrix estimation with high frequency data. J. Econom. 2014; 180: 217–232. van Wieringen WN, Peeters CFW. Ridge estimation of inverse covariance matrices from high-dimensional data. Comput. Statist. Data Anal. 2016; 103: 284–303. Wang L, Chen Z, Wang CD, Li R. Ultrahigh dimensional precision matrix estimation via refitted cross validation. J. Econom. 2020; 215: 118–130. Wang X, Zhang B. Target selection in shrinkage estimation of covariance matrix: A structural similarity approach. Stat. Prob. Lett. 2024; 208: 110048. Dey DK, Srinivasan C. Estimation of a covariance matrix under Stein’s loss. Ann. Stat. 1985; 13: 1581–1591. Ledoit O, Wolf M. A well-conditioned estimator for large-dimensional covariance matrices. J. Multivar. Anal. 2004a; 88: 365–411. Ledoit O, Wolf M. Nonlinear shrinkage estimation of large-dimensional covariance matrices. Ann. Statist. 2012b; 40: 1024–1060. Bodnar T, Gupta AK, Parolya N. Direct shrinkage estimation of large dimensional precision matrix. J. Multivar Anal, 2016; 146: 223–236. Engel J, Buydens L, Blanchet L. An overview of large-dimensional covariance and precision matrix estimators with applications in chemometrics. J. Chemometrics. 2017; 31: e2880. Gray H, Leday GGR, Vallejos CA, Richardson S. Shrinkage estimation of large covariance matrices using multiple shrinkage targets. arXiv preprint. 2018; 1809: 08024. Bartz D, Höhne J, Müller KR. Multi-target shrinkage. arXiv preprint. 2014; 1412: 2041. Lancewicki T, Aladjem M. Multi-target shrinkage estimation for covariance matrices. IEEE Trans. Signal Process. 2014; 62: 6380–6390. Robbins H. Some thoughts on empirical Bayes estimation. Ann. Stat. 1983; 11: 713–723. Bai J, Shi S. Estimating high dimensional covariance matrices and its applications. Ann. Econom. Finance. 2011; 12: 199–215. Chen CF. Bayesian inference for a normal dispersion matrix and its application to stochastic multiple regression analysis. J. Royal Stat. Soc. Ser. B Stat. Methodol. 1979; 41: 235–248. Ledoit O, Wolf M. Improved estimation of the covariance matrix of stock returns with an application to portfolio selection. J. Empir. Finance. 2003; 10: 603–621. Schäfer J, Strimmer K. A shrinkage approach to large-scale covariance matrix estimation and implications for functional genomics. Stat. Appl. Genet. Mol. Biol. 2005; 4: 32. Dobra A, Hans C, Jones B, et al. Sparse graphical models for exploring gene expression data. J. Multivar. Anal. 2014; 90: 196–212. Friedman JH. Regularized discriminant analysis. J. Amer. Statist. Assoc. 1989; 84: 165–175. Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning. Springer Publishing; 2001. Ledoit O, Wolf M. Honey, I shrunk the sample covariance matrix. J. Portfolio. Management. 2024c; 30: 110–119. Ledoit O, Wolf M. Analytical nonlinear shrinkage of large-dimensional covariance matrices. Ann. Stat. 2020d; 48: 3043–3065. Bodnar T, Gupta AK, Parolya N. On the strong convergence of the optimal linear shrinkage estimator for large dimensional covariance matrix. J. Multivar. Anal. 2014; 132: 215–228. Amenc N, Sourd VL. Portfolio Theory and Performance Analysis. John Wiley & Sons Publishing; 2005. Cai T, Hu J, Li Y, Zheng X. High-dimensional minimum variance portfolio estimation based on high-frequency data. J. Econom. 2019; 214: 482–494. Ding Y, Li Y, Zheng X. High dimensional minimum variance portfolio estimation under statistical factor models. J. Econom. 2021; 222: 502–515. Joo YC, Park SY. Optimal portfolio selection using a simple double-shrinkage selection rule. Finance Res. Lett. 2021; 43: 102019.



DOI: https://doi.org/10.24294/fsj9912

Refbacks

  • There are currently no refbacks.


Copyright (c) 2024 Yuang Xue, Lihong Wang

License URL: https://creativecommons.org/licenses/by/4.0/