{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T16:58:43Z","timestamp":1791392323303,"version":"4.3.3"},"reference-count":43,"publisher":"Institute of Mathematical Statistics","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Ann. Statist."],"published-print":{"date-parts":[[2025,8,1]]},"abstract":"<jats:p>Covariance regression analysis is an approach to linking the covariance of responses to a set of explanatory variables X, where X can be a vector, matrix, or tensor. Most of the literature on this topic focuses on the \u201cFixed-X\u201d setting and treats X as nonrandom. By contrast, treating explanatory variables X as random, namely the \u201cRandom-X\u201d setting, is often more realistic in practice. This article aims to fill this gap in the literature on the estimation and model assessment theory for Random-X covariance regression models. Specifically, we construct a new theoretical framework for studying the covariance estimators under the Random-X setting, and we demonstrate that the quasi-maximum likelihood estimator and the weighted least squares estimator are both consistent and asymptotically normal. In addition, we develop pioneering work on the model assessment theory of covariance regression. In particular, we obtain the bias-variance decompositions for the expected test errors under both the Fixed-X and Random-X settings. We show that moving from a Fixed-X to a Random-X setting can increase both the bias and the variance in expected test errors. Subsequently, we propose the estimators of the expected test errors under the Fixed-X and Random-X settings, which can be used to assess the performance of the competing covariance regression models. The proposed estimation and model assessment approaches are illustrated via extensive simulation experiments and an empirical study of stock returns in the US market.<\/jats:p>","DOI":"10.1214\/25-aos2515","type":"journal-article","created":{"date-parts":[[2025,9,12]],"date-time":"2025-09-12T05:22:32Z","timestamp":1757654552000},"page":"1587-1612","source":"Crossref","is-referenced-by-count":2,"title":["Fixed and random covariance regression analyses"],"prefix":"10.1214","volume":"53","author":[{"given":"Tao","family":"Zou","sequence":"first","affiliation":[{"name":"Research School of Finance, Actuarial Studies and Statistics, The Australian National University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Lan","sequence":"additional","affiliation":[{"name":"School of Statistics and Center of Statistical Research, Southwestern University of Finance and Economics"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Runze","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Statistics, The Pennsylvania State University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chih-Ling","family":"Tsai","sequence":"additional","affiliation":[{"name":"Graduate School of Management, University of California, Davis"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"108","reference":[{"key":"12","unstructured":"<small>FOX, E. B.<\/small> and <small>DUNSON, D. B.<\/small> (2015). Bayesian nonparametric covariance regression. <i>J. Mach. Learn. Res.<\/i> <b>16<\/b> 2501\u20132542."},{"key":"23","doi-asserted-by":"crossref","unstructured":"<small>MARKOWITZ, H.<\/small> (1952). Portfolio selection. <i>J. Finance<\/i> <b>7<\/b> 77\u201391.","DOI":"10.1111\/j.1540-6261.1952.tb01525.x"},{"key":"33","unstructured":"<small>Vapnik, V. N.<\/small> (1998). <i>Statistical Learning Theory<\/i>. <i>Adaptive and Learning Systems for Signal Processing, Communications, and Control<\/i>. Wiley, New York."},{"key":"24","unstructured":"<small>MCCULLOCH, C. E.<\/small>, <small>SEARLE, S. R.<\/small> and <small>NEUHAUS, J. M.<\/small> (2008). <i>Generalized, Linear, and Mixed Models<\/i>, 2nd ed. <i>Wiley Series in Probability and Statistics<\/i>. Wiley, Hoboken, NJ."},{"key":"9","unstructured":"<small>DEMIDENKO, E.<\/small> (2013). <i>Mixed Models<\/i>: <i>Theory and Applications with R<\/i>, 2nd ed. <i>Wiley Series in Probability and Statistics<\/i>. Wiley, Hoboken, NJ."},{"key":"32","doi-asserted-by":"publisher","unstructured":"<small>van der Vaart, A. W.<\/small> (1998). <i>Asymptotic Statistics<\/i>. <i>Cambridge Series in Statistical and Probabilistic Mathematics<\/i> <b>3<\/b>. Cambridge Univ. Press, Cambridge.","DOI":"10.1017\/CBO9780511802256"},{"key":"13","doi-asserted-by":"publisher","unstructured":"<small>Gy\u00f6rfi, L.<\/small>, <small>Kohler, M.<\/small>, <small>Krzy\u017cak, A.<\/small> and <small>Walk, H.<\/small> (2002). <i>A Distribution-Free Theory of Nonparametric Regression<\/i>. <i>Springer Series in Statistics<\/i>. Springer, New York.","DOI":"10.1007\/b97848"},{"key":"27","doi-asserted-by":"publisher","unstructured":"<small>Rosset, S.<\/small> and <small>Tibshirani, R. J.<\/small> (2020). From Fixed-X to Random-X regression: Bias-variance decompositions, covariance penalties, and prediction error estimation. <i>J. Amer. Statist. Assoc.<\/i> <b>115<\/b> 138\u2013151.","DOI":"10.1080\/01621459.2018.1424632"},{"key":"2","doi-asserted-by":"publisher","unstructured":"<small>Bai, Z. D.<\/small> and <small>Silverstein, J. W.<\/small> (2004). CLT for linear spectral statistics of large-dimensional sample covariance matrices. <i>Ann. Probab.<\/i> <b>32<\/b> 553\u2013605.","DOI":"10.1214\/aop\/1078415845"},{"key":"25","doi-asserted-by":"publisher","unstructured":"<small>Pan, G. M.<\/small> and <small>Zhou, W.<\/small> (2008). Central limit theorem for signal-to-interference ratio of reduced rank linear receiver. <i>Ann. Appl. Probab.<\/i> <b>18<\/b> 1232\u20131270.","DOI":"10.1214\/07-AAP477"},{"key":"39","doi-asserted-by":"publisher","unstructured":"<small>Zheng, S.<\/small> (2012). Central limit theorems for linear spectral statistics of large dimensional <i>F<\/i>-matrices. <i>Ann. Inst. Henri Poincar\u00e9 Probab. Stat.<\/i> <b>48<\/b> 444\u2013476.","DOI":"10.1214\/11-AIHP414"},{"key":"14","doi-asserted-by":"publisher","unstructured":"<small>Hastie, T.<\/small>, <small>Tibshirani, R.<\/small> and <small>Friedman, J.<\/small> (2009). <i>The Elements of Statistical Learning<\/i>: <i>Data Mining, Inference, and Prediction<\/i>, 2nd ed. <i>Springer Series in Statistics<\/i>. Springer, New York.","DOI":"10.1007\/978-0-387-84858-7"},{"key":"35","doi-asserted-by":"publisher","unstructured":"<small>White, H.<\/small> (1982). Maximum likelihood estimation of misspecified models. <i>Econometrica<\/i> <b>50<\/b> 1\u201325.","DOI":"10.2307\/1912526"},{"key":"8","doi-asserted-by":"publisher","unstructured":"<small>Cressie, N. A. C.<\/small> (1993). <i>Statistics for Spatial Data<\/i>. <i>Wiley Series in Probability and Mathematical Statistics<\/i>: <i>Applied Probability and Statistics<\/i>. Wiley, New York.","DOI":"10.1002\/9781119115151"},{"key":"7","unstructured":"<small>CHUNG, K. L.<\/small> (2001). <i>A Course in Probability Theory<\/i>, 3rd ed. Academic Press, San Diego, CA."},{"key":"1","doi-asserted-by":"crossref","unstructured":"<small>ANDERSON, T. W.<\/small> (1973). Asymptotically efficient estimation of covariance matrices with linear structure. <i>Ann. Statist.<\/i> <b>1<\/b> 135\u2013141.","DOI":"10.1214\/aos\/1193342389"},{"key":"6","doi-asserted-by":"publisher","unstructured":"<small>CHIU, T. Y. M.<\/small>, <small>LEONARD, T.<\/small> and <small>TSUI, K.-W.<\/small> (1996). The matrix-logarithmic covariance model. <i>J. Amer. Statist. Assoc.<\/i> <b>91<\/b> 198\u2013210.","DOI":"10.2307\/2291396"},{"key":"15","doi-asserted-by":"publisher","unstructured":"<small>HOFF, P. D.<\/small> and <small>NIU, X.<\/small> (2012). A covariance regression model. <i>Statist. Sinica<\/i> <b>22<\/b> 729\u2013753.","DOI":"10.5705\/ss.2010.051"},{"key":"41","doi-asserted-by":"publisher","unstructured":"<small>ZOU, T.<\/small>, <small>LAN, W.<\/small>, <small>LI, R.<\/small> and <small>TSAI, C.-L.<\/small> (2022). Inference on covariance-mean regression. <i>J. Econometrics<\/i> <b>230<\/b> 318\u2013338.","DOI":"10.1016\/j.jeconom.2021.05.004"},{"key":"43","doi-asserted-by":"publisher","unstructured":"<small>ZOU, T.<\/small>, <small>LAN, W.<\/small>, <small>WANG, H.<\/small> and <small>TSAI, C.-L.<\/small> (2017). Covariance regression analysis. <i>J. Amer. Statist. Assoc.<\/i> <b>112<\/b> 266\u2013281.","DOI":"10.1080\/01621459.2015.1131699"},{"key":"31","doi-asserted-by":"publisher","unstructured":"<small>TSE, Y. K.<\/small> and <small>TSUI, A. K. C.<\/small> (2002). A multivariate generalized autoregressive conditional heteroscedasticity model with time-varying correlations. <i>J. Bus. Econom. Statist.<\/i> <b>20<\/b> 351\u2013362.","DOI":"10.1198\/073500102288618496"},{"key":"3","doi-asserted-by":"publisher","unstructured":"<small>BOLLERSLEV, T.<\/small> and <small>WOOLDRIDGE, J. M.<\/small> (1992). Quasi-maximum likelihood estimation and inference in dynamic models with time-varying covariances. <i>Econometric Rev.<\/i> <b>11<\/b> 143\u2013172.","DOI":"10.1080\/07474939208800229"},{"key":"4","doi-asserted-by":"crossref","unstructured":"<small>BREIMAN, L.<\/small> and <small>SPECTOR, P.<\/small> (1992). Submodel selection and evaluation in regression. The X-random case. <i>Int. Stat. Rev.<\/i> <b>60<\/b> 291\u2013319.","DOI":"10.2307\/1403680"},{"key":"5","doi-asserted-by":"publisher","unstructured":"<small>CAI, T. T.<\/small>, <small>HU, J.<\/small>, <small>LI, Y.<\/small> and <small>ZHENG, X.<\/small> (2020). High-dimensional minimum variance portfolio estimation based on high-frequency data. <i>J. Econometrics<\/i> <b>214<\/b> 482\u2013494.","DOI":"10.1016\/j.jeconom.2019.04.039"},{"key":"10","doi-asserted-by":"crossref","unstructured":"<small>DEMIGUEL, V.<\/small>, <small>GARLAPPI, L.<\/small> and <small>UPPAL, R.<\/small> (2009). Optimal versus naive diversification: How inefficient is the 1\/N portfolio strategy? <i>Rev. Financ. Stud.<\/i> <b>22<\/b> 1915\u20131953.","DOI":"10.1093\/rfs\/hhm075"},{"key":"11","doi-asserted-by":"publisher","unstructured":"<small>FAN, J.<\/small>, <small>LI, Y.<\/small> and <small>YU, K.<\/small> (2012). Vast volatility matrix estimation using high-frequency data for portfolio selection. <i>J. Amer. Statist. Assoc.<\/i> <b>107<\/b> 412\u2013428.","DOI":"10.1080\/01621459.2012.656041"},{"key":"16","doi-asserted-by":"publisher","unstructured":"<small>HUANG, D.<\/small>, <small>LAN, W.<\/small>, <small>ZHANG, H. H.<\/small> and <small>WANG, H.<\/small> (2019). Least squares estimation of spatial autoregressive models for large-scale social networks. <i>Electron. J. Stat.<\/i> <b>13<\/b> 1135\u20131165.","DOI":"10.1214\/19-ejs1549"},{"key":"17","doi-asserted-by":"publisher","unstructured":"<small>KELEJIAN, H. H.<\/small> and <small>PRUCHA, I. R.<\/small> (2001). On the asymptotic distribution of the Moran <i>I<\/i> test statistic with applications. <i>J. Econometrics<\/i> <b>104<\/b> 219\u2013257.","DOI":"10.1016\/S0304-4076(01)00064-1"},{"key":"18","doi-asserted-by":"publisher","unstructured":"<small>LAURENT, S.<\/small>, <small>ROMBOUTS, J. V. K.<\/small> and <small>VIOLANTE, F.<\/small> (2012). On the forecasting accuracy of multivariate GARCH models. <i>J. Appl. Econometrics<\/i> <b>27<\/b> 934\u2013955.","DOI":"10.1002\/jae.1248"},{"key":"20","doi-asserted-by":"crossref","unstructured":"<small>LESAGE, J.<\/small> and <small>PACE, R. K.<\/small> (2009). <i>Introduction to Spatial Econometrics<\/i>. CRC Press, New York.","DOI":"10.1201\/9781420064254"},{"key":"21","doi-asserted-by":"crossref","unstructured":"<small>MA, Y.<\/small>, <small>PAN, R.<\/small>, <small>ZOU, T.<\/small> and <small>WANG, H.<\/small> (2020). A naive least squares method for spatial autoregression with covariates. <i>Statist. Sinica<\/i> <b>30<\/b> 653\u2013672.","DOI":"10.5705\/ss.202017.0135"},{"key":"22","unstructured":"<small>MALLOWS, C. L.<\/small> (1973). Some comments on <math xmlns=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" overflow=\"scroll\" id=\"eq_1\"><msub><mrow><mi mathvariant=\"italic\">C<\/mi><\/mrow><mrow><mi mathvariant=\"italic\">p<\/mi><\/mrow><\/msub><\/math>. <i>Technometrics<\/i> <b>15<\/b> 661\u2013675."},{"key":"26","doi-asserted-by":"publisher","unstructured":"<small>POIGNARD, B.<\/small> and <small>ASAI, M.<\/small> (2023). High-dimensional sparse multivariate stochastic volatility models. <i>J. Time Series Anal.<\/i> <b>44<\/b> 4\u201322.","DOI":"10.1111\/jtsa.12647"},{"key":"28","doi-asserted-by":"crossref","unstructured":"<small>SCOTT, M. A.<\/small> and <small>HANDCOCK, M. S.<\/small> (2001). Covariance models for latent structure in longitudinal data. <i>Sociol. Method.<\/i> <b>31<\/b> 265\u2013303.","DOI":"10.1111\/0081-1750.00097"},{"key":"29","doi-asserted-by":"publisher","unstructured":"<small>SHEN, X.<\/small> and <small>HUANG, H.-C.<\/small> (2020). Discussion of \u201cFrom Fixed-X to Random-X regression: Bias-variance decompositions, covariance penalties, and prediction error estimation\u201d [ MR4078450]. <i>J. Amer. Statist. Assoc.<\/i> <b>115<\/b> 152\u2013156.","DOI":"10.1080\/01621459.2018.1543597"},{"key":"30","unstructured":"<small>TSAY, R. S.<\/small> (2014). <i>Multivariate Time Series Analysis<\/i>. <i>Wiley Series in Probability and Statistics<\/i>. Wiley, Hoboken, NJ."},{"key":"36","doi-asserted-by":"crossref","unstructured":"<small>WIKLE, C. K.<\/small>, <small>ZAMMIT-MANGION, A.<\/small> and <small>CRESSIE, N.<\/small> (2019). <i>Spatio-Temporal Statistics with R<\/i>. CRC Press, Boca Raton, FL.","DOI":"10.1201\/9781351769723"},{"key":"37","unstructured":"<small>WOOLDRIDGE, J. M.<\/small> (2002). <i>Econometric Analysis of Cross Section and Panel Data<\/i>. MIT Press, Cambridge."},{"key":"38","unstructured":"<small>WOOLDRIDGE, J. M.<\/small> (2012). <i>Introductory Econometrics<\/i>: <i>A Modern Approach<\/i>. Cengage Learning, Mason, OH."},{"key":"42","doi-asserted-by":"crossref","unstructured":"<small>ZOU, T.<\/small>, <small>LAN, W.<\/small>, <small>LI, R.<\/small> and <small>TSAI, C.-L.<\/small> (2025). Supplement to \u201cFixed and Random Covariance Regression Analyses.\u201d <a href=\"https:\/\/doi.org\/10.1214\/25-AOS2515SUPPA\">https:\/\/doi.org\/10.1214\/25-AOS2515SUPPA<\/a>, <a href=\"https:\/\/doi.org\/10.1214\/25-AOS2515SUPPB\">https:\/\/doi.org\/10.1214\/25-AOS2515SUPPB<\/a>","DOI":"10.1214\/25-AOS2515SUPPB"},{"key":"34","doi-asserted-by":"publisher","unstructured":"<small>WANG, L.<\/small>, <small>BROWN, L. D.<\/small>, <small>CAI, T. T.<\/small> and <small>LEVINE, M.<\/small> (2008). Effect of mean on variance function estimation in nonparametric regression. <i>Ann. Statist.<\/i> <b>36<\/b> 646\u2013664.","DOI":"10.1214\/009053607000000901"},{"key":"40","doi-asserted-by":"publisher","unstructured":"<small>ZHOU, J.<\/small>, <small>TU, Y.<\/small>, <small>CHEN, Y.<\/small> and <small>WANG, H.<\/small> (2017). Estimating spatial autocorrelation with sampled network data. <i>J. Bus. Econom. Statist.<\/i> <b>35<\/b> 130\u2013138.","DOI":"10.1080\/07350015.2015.1061437"},{"key":"19","doi-asserted-by":"publisher","unstructured":"<small>LEE, L.-F.<\/small> (2004). Asymptotic distributions of quasi-maximum likelihood estimators for spatial autoregressive models. <i>Econometrica<\/i> <b>72<\/b> 1899\u20131925.","DOI":"10.1111\/j.1468-0262.2004.00558.x"}],"container-title":["The Annals of Statistics"],"original-title":[],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T16:37:50Z","timestamp":1791391070000},"score":1,"resource":{"primary":{"URL":"https:\/\/projecteuclid.org\/journals\/annals-of-statistics\/volume-53\/issue-4\/Fixed-and-random-covariance-regression-analyses\/10.1214\/25-AOS2515.full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,1]]},"references-count":43,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,8,1]]}},"URL":"https:\/\/doi.org\/10.1214\/25-aos2515","relation":{},"ISSN":["0090-5364","2168-8966"],"issn-type":[{"value":"0090-5364","type":"print"},{"value":"2168-8966","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,1]]}}}