Structure detection of semiparametric structural equation models with Bayesian adaptive group lasso

被引:26
作者
Feng, Xiang-Nan [1 ]
Wang, Guo-Chang [2 ]
Wang, Yi-Fan [1 ]
Song, Xin-Yuan [1 ]
机构
[1] Chinese Univ Hong Kong, Dept Stat, Shatin, Hong Kong, Peoples R China
[2] Ji Nan Univ, Coll Econ, Guangzhou 510236, Guangdong, Peoples R China
基金
中国博士后科学基金; 中国国家自然科学基金;
关键词
Bayesian adaptive group lasso; basis expansion approximation; MCMC methods; semiparametric SEMs; LATENT VARIABLE MODELS; METABOLIC SYNDROME; RENAL-FUNCTION; REGRESSION; SELECTION; COEFFICIENT; POPULATION; SHRINKAGE;
D O I
10.1002/sim.6410
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
Structural equation models (SEMs) are widely recognized as the most important statistical tool for assessing the interrelationships among latent variables. This study develops a Bayesian adaptive group least absolute shrinkage and selection operator procedure to perform simultaneous model selection and estimation for semiparametric SEMs, wherein the structural equation is formulated using the additive nonparametric functions of observed and latent variables. We propose the use of basis expansions to approximate the unknown functions. By introducing adaptive penalties to the groups of basis expansions, the nonlinear, linear, or non-existent effects of observed and latent variables in the structural equation can be automatically detected. A simulation study demonstrates that the proposed method performs satisfactorily. This paper presents an application of revealing the observed and latent risk factors of diabetic kidney disease. Copyright (c) 2015 John Wiley & Sons, Ltd.
引用
收藏
页码:1527 / 1547
页数:21
相关论文
共 54 条
[1]   Bayesian adaptive Lasso quantile regression [J].
Alhamzawi, Rahim ;
Yu, Keming ;
Benoit, Dries F. .
STATISTICAL MODELLING, 2012, 12 (03) :279-297
[2]   Factor analysis of the metabolic syndrome:: obesity vs insulin resistance as the central abnormality [J].
Anderson, PJ ;
Critchley, JAJH ;
Chan, JCN ;
Cockram, CS ;
Lee, ZSK ;
Thomas, GN ;
Tomlinson, B .
INTERNATIONAL JOURNAL OF OBESITY, 2001, 25 (12) :1782-1788
[3]  
ANDREWS DF, 1974, J ROY STAT SOC B MET, V36, P99
[4]  
[Anonymous], ELEMENTS STAT LEARNI
[5]  
Bach FR, 2008, J MACH LEARN RES, V9, P1179
[6]   Preserving renal function in adults with hypertension and diabetes: A consensus approach [J].
Bakris, GL ;
Williams, M ;
Dworkin, L ;
Elliott, WJ ;
Epstein, M ;
Toto, R ;
Tuttle, K ;
Douglas, J ;
Hsueh, W ;
Sowers, J .
AMERICAN JOURNAL OF KIDNEY DISEASES, 2000, 36 (03) :646-661
[7]   A semiparametric approach to modeling nonlinear relations among latent variables [J].
Bauer, DJ .
STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 2005, 12 (04) :513-535
[8]  
Bollens KA, 1989, Structural Equations With Latent Variables
[9]   Smoothing Population Size Estimates for Time-Stratified Mark-Recapture Experiments Using Bayesian P-Splines [J].
Bonner, Simon J. ;
Schwarz, Carl J. .
BIOMETRICS, 2011, 67 (04) :1498-1507
[10]  
Buhlmanns P, 2011, STAT HIGH DIMENSIONA