Homogeneity Estimation in Multivariate Generalized Linear Models

Abstract

Multivariate regression models have been extensively studied in the literature and applied in practice. It is not unusual that some predictors may make the same nonnull contributions to all the elements of the response vector, especially when the number of predictors is very large. For convenience, we call the set of such predictors as the homogeneity set. In this paper, we consider a sparse high-dimensional multivariate generalized linear models with coexisting homogeneity and heterogeneity sets of predictors, which is very important to facilitate the understanding of the effects of different types of predictors as well as improvement on the estimation efficiency. We propose a novel adaptive regularized method by which we can easily identify the homogeneity set of predictors and investigate the asymptotic properties of the parameter estimation. More importantly, the proposed method yields a smaller variance for parameter estimation compared to the ones that do not consider the existence of a homogeneity set of predictors. We also provide a computational algorithm and present its theoretical justification. In addition, we perform extensive simulation studies and present real data examples to demonstrate the proposed method.

Keywords

Asymptotic variance / Detection consistency / Homogeneity and heterogeneity / Multivariate generalized linear model / 62F12 / 62H12

References

[1] ChenL, HuangJZ. Sparse reduced-rank regression for simultaneous dimension reduction and variable selection. J. Am. Stat. Assoc., 2012, 107: 1533-1545.

[2] ChenS, et al.. MicroRNA in ovarian cancer: biology, pathogenesis, and therapeutic opportunities. Int. J. Environ. Res. Public Health, 2019, 161510.

[3] ChenY, LuoZ, KongL. $\ell _{2,0}$-norm based selection and estimation for multivariate generalized linear models. J. Multivar. Anal., 2021, 185. 104782

[4] CitronM. Alzheimer’s disease: strategies for disease modification. Nat. Rev. Drug Discov., 2010, 9: 387-398.

[5] Echevarria-VargasIM, ValiyevaF, Vivas-MejiaPE. Upregulation of miR-21 in cisplatin resistant ovarian cancer via JNK-1/c-Jun pathway. PLoS One, 2014, 9. e97094

[6] FahrmeirL, KanfmannH. Consistency and asymptotic normality of the maximum likelihood estimator in generalized linear models. Ann. Stat., 1985, 13: 342-368.

[7] FanJ, LvJ. Nonconcave penalized likelihood with NP-dimensionality. IEEE Trans. Inf. Theory, 2011, 57: 5467-5484.

[8] GaoYC, WuJ. MicroRNA-200c and microRNA-141 as potential diagnostic and prognostic biomarkers for ovarian cancer. Tumour Biol., 2015, 36: 4843-4850.

[9] HuX, et al.. A miR-200 microRNA cluster as prognostic marker in advanced ovarian cancer. Gynecol. Oncol., 2009, 114: 457-464.

[10] HuaX, et al.. Sex and age differences in atrophic rates: an ADNI study with $n=1368$ MRI scans. Neurobiol. Aging, 2010, 31: 1463-1480.

[11] IorioMV, et al.. MicroRNA signatures in human ovarian cancer. Cancer Res., 2007, 67: 8699-8707.

[12] KangK, CaiJH, SongXY, ZhuHT. Bayesian hidden Markov models for delineating the pathology of Alzheimer’s disease. Stat. Methods Med. Res., 2019, 28: 2112-2124.

[13] LanH, ChenW, HeG, YangS. miR-140-5p inhibits ovarian cancer growth partially by repression of PDGFRA. Biomed. Pharmacother., 2015, 75: 117-122.

[14] LiX, et al.. miR-142-5p enhances cisplatin-induced apoptosis in ovarian cancer cells by targeting multiple anti-apoptotic genes. Biochem. Pharmacol., 2019, 161: 98-112.

[15] LiY, NanB, ZhuJ. Multivariate sparse group lasso for the multivariate multiple linear regression with an arbitrary group structure. Biometrics, 2015, 71: 354-363.

[16] LuoP, FeiJ, ZhouJ, ZhangW. microRNA-126 suppresses PAK4 expression in ovarian cancer SKOV3 cells. Oncol. Lett., 2015, 9: 2225-2229.

[17] MccullaghP, NelderJAGeneralized linear models, 19892London. Chapman & Hall. .

[18] NelderJA, WedderburnRWM. Generalized linear models. J. R. Stat. Soc. Ser. A, 1972, 135: 370-384.

[19] ObozinskiG, WainwrightMJ, JordanMI. Support union recovery in high-dimensional multivariate regression. Ann. Stat., 2011, 39: 1-47.

[20] ReyAL’examen clinique en psychologie, 1958, Paris. Presses Universitaries de France.

[21] RosenW, MohsRC, DavisKL. A new rating scale for Alzheimer’s disease. Am. J. Psychiatry, 1984, 141: 1356-1364.

[22] SammelM, LinX, RyanL. Multivariate linear mixed models for multiple outcomes. Stat. Med., 1999, 18: 2479-2492.

[23] SchaidDJ, TongX, BatzlerA, SinnwellJP, QingJ, BiernackaJM. Dimension reduction and coefficient estimation in multivariate linear regression. Biostatistics, 2019, 20: 111-128

[24] ShenX, HuangH, PanW. Simultaneous supervised clustering and feature selection over a graph. Biometrika, 2012, 99: 899-914.

[25] ShiC, SongR, ChenZ, LiR. Linear hypothesis testing for high dimensional generalized linear models. Ann. Stat., 2019, 47: 2671-2703.

[26] The Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature, 2011, 474: 609-615.

[27] TsengP. Convergence of a block coordinate descent method for nondifferentiable minimization. J. Optim. Theory Appl., 2001, 109: 475-494.

[28] TurlachBA, VenablesWN, WrightSJ. Simultaneous variable selection. Technometrics, 2005, 47: 349-363.

[29] WangX, FanJ. Variable selection for multivariate generalized linear models. J. Appl. Stat., 2014, 41: 393-406.

[30] WangZ, LiuX, TangW, LinY. Incorporating graphical structure of predictors in sparse quantile regression. J. Bus. Econ. Stat., 2021, 39: 783-792.

[31] WeiJ, et al.. MicroRNA-205 promotes cell invasion by repressing TCF21 in human ovarian cancer. J. Ovarian Res., 2017, 1033.

[32] YangH, et al.. MicroRNA expression profiling in human ovarian cancer: miR-214 induces cell survival and cisplatin resistance by targeting PTEN. Cancer Res., 2008, 68: 425-433.

[33] YuanM, EkiciA, LuZ, MonteiroR. Dimension reduction and coefficient estimation in multivariate linear regression. J. R. Stat. Soc. Ser. B, 2007, 69: 329-346.

[34] YuanM, LinY. Model selection and estimation in regression with grouped variables. J. R. Stat. Soc. Ser. B, 2006, 68: 49-67.

[35] ZhangL, et al.. Genomic and epigenetic alterations deregulate microRNA expression in human epithelial ovarian cancer. Proc. Natl. Acad. Sci. USA, 2008, 105: 7004-7009.

[36] ZouH. The adaptive lasso and its oracle properties. J. Am. Stat. Assoc., 2006, 101: 1418-1429.

[37] ZouC, KeY, ZhangW. Estimation of low rank high-dimensional multivariate linear models for multi-response data. J. Am. Stat. Assoc., 2020.

Funding

National Natural Science Foundation of China(11971457)

University Natural Science Research Project of Anhui Province(1908085MA06)

http://dx.doi.org/10.13039/501100012226(WK2040000035)

Natural Sciences and Engineering Research Council of Canada(RGPIN-2017-05720)

RIGHTS & PERMISSIONS

School of Mathematical Sciences, University of Science and Technology of China and Springer-Verlag GmbH Germany, part of Springer Nature