A majorization algorithm for simultaneous parameter estimation in robust exploratory factor analysis

A new approach for fitting the exploratory factor analysis (EFA) model is considered. The EFA model is fitted directly to the data matrix by minimizing a weighted least squares (WLS) goodness-of-fit measure. The WLS fitting problem is solved by iteratively performing unweighted least squares fitting of the same model. A convergent reweighted least squares algorithm based on iterative majorization is developed. The influence of large residuals in the loss function is curbed using Huber's criterion. This procedure leads to robust EFA that can resist the effect of outliers in the data. Applications to real and simulated data illustrate the performance of the proposed approach.

[1]  George A. Marcoulides,et al.  Latent variable and latent structure models , 2002 .

[2]  S. Zamir,et al.  Lower Rank Approximation of Matrices by Least Squares With Any Choice of Weights , 1979 .

[3]  Peter J. Rousseeuw,et al.  Robust regression and outlier detection , 1987 .

[4]  Frederick Mosteller,et al.  Data Analysis and Regression , 1978 .

[5]  Johan H. L. Oud,et al.  Recent developments on structural equations models: theory and applications , 2004 .

[6]  Kenneth Lange,et al.  Sharp quadratic majorization in one dimension , 2009, Comput. Stat. Data Anal..

[7]  Patrick J. F. Groenen,et al.  New developments in classification and data analysis , 2005 .

[8]  W. J. Krzanowski,et al.  Recent Advances in Descriptive Multivariate Analysis. , 1996 .

[9]  R. Kosfeld.  Robust exploratory factor analysis , 1996 .

[10]  Peter Filzmoser,et al.  Regressions , 2019, Energy Transfers by Convection.

[11]  Richard A. Johnson,et al.  Applied Multivariate Statistical Analysis , 1983 .

[12]  H. Harman.  Modern factor analysis , 1961 .

[13]  Nickolay T. Trendafilov,et al.  Exploratory Factor Analysis of Data Matrices With More Variables Than Observations , 2011 .

[14]  S. Mulaik,et al.  Foundations of Factor Analysis , 1975 .

[15]  P. Filzmoser.  Robust principal component and factor analysis in the geostatistical treatment of environmental data , 1999 .

[16]  Peter Verboon,et al.  A robust approach to nonlinear multivariate analysis , 1994 .

[17]  Willem J. Heiser,et al.  Resistant orthogonal procrustes analysis , 1992 .

[18]  Ursula Gather,et al.  The largest nonindentifiable outlier: a comparison of multivariate simultaneous outlier identification rules , 2001 .

[19]  Hans-Hermann Bock,et al.  Information Systems and Data Analysis , 1994 .

[20]  Jan de Leeuw,et al.  Block-relaxation Algorithms in Statistics , 1994 .

[21]  P. Groenen,et al.  An improved majorization algorithm for robust procrustes analysis. , 2005 .

[22]  P. Rousseeuw,et al.  Robust factor analysis , 2003 .

[23]  P. Rousseeuw.  Multivariate estimation with high breakdown point , 1985 .

[24]  Patrick J. F. Groenen,et al.  Weighted Majorization Algorithms for Weighted Least Squares Decomposition Models , 2003 .

[25]  Albert Maydeu-Olivares,et al.  Contemporary psychometrics : a festschrift for Roderick P. McDonald , 2005 .

[26]  J. Leeuw,et al.  Least Squares Optimal Scaling of Partially Observed Linear Systems , 2003 .

[27]  H. Kiers.  Weighted least squares fitting using ordinary least squares algorithms , 1997 .

[28]  Dimitris Mavridis,et al.  Detecting Outliers in Factor Analysis Using the Forward Search Algorithm , 2008, Multivariate behavioral research.