Optimizing Ridge Generalized Least Squares for Structural Equation Modeling

被引:13
作者
Yang, Miao [1 ]
Yuan, Ke-Hai [1 ]
机构
[1] Univ Notre Dame, Notre Dame, IN USA
基金
美国国家科学基金会;
关键词
efficiency; empirical modeling; Monte Carlo simulation; parameter estimation; TEST STATISTICS; FIT INDEXES; CONVERGENCE; SELECTION; SKEWNESS; KURTOSIS; PATHS;
D O I
10.1080/10705511.2018.1479853
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Ridge generalized least squares (RGLS) is a recently proposed estimation procedure for structural equation modeling. In the formulation of RGLS, there is a key element, ridge tuning parameter, whose value determines the efficiency of parameter estimates. This article aims to optimize RGLS by developing formulas for the ridge tuning parameter to yield the most efficient parameter estimates in practice. For the formulas to have a wide scope of applicability, they are calibrated using empirical efficiency and via many conditions on population distribution, sample size, number of variables, and model structure. Results show that RGLS with the tuning parameter determined by the formulas can substantially improve the efficiency of parameter estimates over commonly used procedures with real data being typically nonnormally distributed.
引用
收藏
页码:24 / 38
页数:15
相关论文
共 23 条
[1]   THE EFFECT OF SAMPLING ERROR ON CONVERGENCE, IMPROPER SOLUTIONS, AND GOODNESS-OF-FIT INDEXES FOR MAXIMUM-LIKELIHOOD CONFIRMATORY FACTOR-ANALYSIS [J].
ANDERSON, JC ;
GERBING, DW .
PSYCHOMETRIKA, 1984, 49 (02) :155-173
[2]  
[Anonymous], 1982, ROBUSTNESS LISREL SM
[3]   Skewness and Kurtosis in Real Data Samples [J].
Blanca, Maria J. ;
Arnau, Jaume ;
Lopez-Montiel, Dolores ;
Bono, Roser ;
Bendayan, Rebecca .
METHODOLOGY-EUROPEAN JOURNAL OF RESEARCH METHODS FOR THE BEHAVIORAL AND SOCIAL SCIENCES, 2013, 9 (02) :78-84
[5]   When fit indices and residuals are incompatible [J].
Browne, MW ;
MacCallum, RC ;
Kim, CT ;
Andersen, BL ;
Glaser, R .
PSYCHOLOGICAL METHODS, 2002, 7 (04) :403-421
[6]   Univariate and multivariate skewness and kurtosis for measuring nonnormality: Prevalence, influence and estimation [J].
Cain, Meghan K. ;
Zhang, Zhiyong ;
Yuan, Ke-Hai .
BEHAVIOR RESEARCH METHODS, 2017, 49 (05) :1716-1735
[7]   Tabu search model selection in multiple regression analysis [J].
Drezner, Z ;
Marcoulides, GA ;
Salhi, S .
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 1999, 28 (02) :349-367
[8]   Regularization Paths for Generalized Linear Models via Coordinate Descent [J].
Friedman, Jerome ;
Hastie, Trevor ;
Tibshirani, Rob .
JOURNAL OF STATISTICAL SOFTWARE, 2010, 33 (01) :1-22
[9]   FUTURE PATHS FOR INTEGER PROGRAMMING AND LINKS TO ARTIFICIAL-INTELLIGENCE [J].
GLOVER, F .
COMPUTERS & OPERATIONS RESEARCH, 1986, 13 (05) :533-549
[10]   CAN TEST STATISTICS IN COVARIANCE STRUCTURE-ANALYSIS BE TRUSTED [J].
HU, LT ;
BENTLER, PM ;
KANO, Y .
PSYCHOLOGICAL BULLETIN, 1992, 112 (02) :351-362