Robust Estimation of Structural Equation Modeling using Mahalanobis Distance-based Trimming: An Application to Job Performance Data

Zulfiqar, Ammara and Aziz, Mahwish and Wahid, Abdul (2026): Robust Estimation of Structural Equation Modeling using Mahalanobis Distance-based Trimming: An Application to Job Performance Data.

Abstract

Structural Equation Modeling (SEM) is a commonly used and prevalent method to describe the relationships between latent and observed variables. If these variables contain outliers and leverage-points, the estimation by existing SEM is problematic and leads to biased and inefficient estimators. In this article, we propose the Least Mahalanobis Distance-based Trimmed (LMDT) model which uses Mahalanobis distance for the identification of outliers in SEM and trimming approach for dealing with such types of influential observations. By using this suggested technique, instead of maximum likelihood and least squares criteria, the LMDT is resistant to outliers in both measurement error and latent factors. A FAST-iterative algorithm is constructed and implemented for computing the LMDT. Both a simulation study and a real data analysis indicate that the proposed robust method has good performance in terms of bias and efficiency on contaminated and non-normal skewed data and it outperforms the two non-robust and one robust existing estimation methods.

• Item Type: MPRA Paper
• Language: English
• Keywords: Structural equation modeling; outliers; non-normality; Mahalanobis distance; trimming
• Subjects: C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C15 - Statistical Simulation Methods: General; C - Mathematical and Quantitative Methods > C5 - Econometric Modeling > C51 - Model Construction and Estimation; J - Labor and Demographic Economics > J2 - Demand and Supply of Labor > J28 - Safety ; Job Satisfaction ; Related Public Policy

References

1. Agulló, J., Croux, C., & Van Aelst, S. (2006). The multivariate least-trimmed squares estimator. Journal of Multivariate Analysis, 99(3), 311-338. https://doi.org/10.1177/002224377701400209

2. Aitkin, M. A. (2012). Simultaneous inference and the choice of variable subsets in multiple regression. Technometrics, 16(2), 221-227. https://doi.org/10.1080/00401706.1974.10489177

3. Asparouhov, T., & Muthén, B. (2015). Structural equation models and mixture models with continuous nonnormal skewed distributions. Structural Equation Modeling: A Multidisciplinary Journal, 23(1), 1-19. https://doi.org/10.1080/10705511.2014.947375

4. Bagozzi, R. P. (1977). Structural equation models in experimental research. Journal of Marketing Research, 14(2), 209-226. https://doi.org/10.1177/002224377701400209

5. Campbell, N. A. (1980). Robust procedures in multivariate analysis I: Robust covariance estimation. Journal of the Royal Statistical Society Series C: Applied Statistics, 29(3), 231-237

6. Khan, D. M., Yaqoob, A., Iqbal, N., Wahid, A., Khalil, U., Khan, M., ... & Khan, Z. (2019). Variable Selection via SCAD-Penalized Quantile Regression for High-Dimensional Count Data. IEEE Access, 7, 153205-153216.

7. Du, H., & Bentler, P. M. (2022). Distributionally weighted least squares in structural equation modeling. Psychological Methods, 27(4), 519. https://doi.org/10.1037/met0000388

8. Fan, X., Thompson, B., & Wang, L. (2009). Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes. Structural equation modeling: a multidisciplinary journal, 6(1), 56-83. https://doi.org/10.1080/10705519909540119

9. Fernández, C., & Steel, M. F. (2012). On Bayesian modeling of fat tails and skewness. Journal of the american statistical association, 93(441), 359-371. https://doi.org/10.1080/01621459.1998.10474117

10. Grotzinger, A. D., Rhemtulla, M., de Vlaming, R., Ritchie, S. J., Mallard, T. T., Hill, W. D., . . . Deary, I. J. (2019). Genomic structural equation modelling provides insights into the multivariate genetic architecture of complex traits. Nature human behaviour, 3(5), 513-525. https://doi.org/10.1038/s41562-019-0566-x

11. Hox, J. J., & Maas, C. J. (2009). The accuracy of multilevel structural equation modeling with pseudobalanced groups and small samples. Structural equation modeling, 8(2), 157-174. https://doi.org/10.1207/S15328007SEM0802_1

12. Ibrahim, O. S., & Mohammed, M. J. (2021). A proposed method for cleaning data from outlier values using the robust RFCH method in structural equation modeling. International Journal of Nonlinear Analysis and Applications, 12(2), 2269-2293. 10.22075/IJNAA.2021.5374

13. Lai, M. H., & Zhang, J. (2017). Evaluating fit indices for multivariate t-based structural equation modeling with data contamination. Frontiers in Psychology, 8, 275593. https://doi.org/10.3389/fpsyg.2017.01286

14. Mindrila, D. (2010). Maximum likelihood (ML) and diagonally weighted least squares (DWLS) estimation procedures: A comparison of estimation bias with ordinal and multivariate non-normal data. International Journal of Digital Society, 1(1), 60-66

15. Muthén, B., & Satorra, A. (2014). Multilevel aspects of varying parameters in structural models Multilevel analysis of educational data (pp. 87-99): Elsevier.

16. Poon, W. Y., Lew, S. F., & Poon, Y. S. (2010). A local influence approach to identifying multiple multivariate outliers. British Journal of Mathematical and Statistical Psychology, 53(2), 255-273. https://doi.org/10.1348/000711000159321

17. Ringle, C. M., Sarstedt, M., Mitchell, R., & Gudergan, S. P. (2018). Partial least squares structural equation modeling in HRM research. The international journal of human resource management, 31(12), 1617-1643. https://doi.org/10.1080/09585192.2017.1416655

18. Rousseeuw, P. J., & Leroy, A. M. (2005). Robust regression and outlier detection: John wiley & sons.

19. Shevlin, M., & Miles, J. N. (1999). Effects of sample size, model specification and factor loadings on the GFI in confirmatory factor analysis. Personality and Individual differences, 25(1), 85-90. https://doi.org/10.1016/S0191-8869(98)00055-5

20. Steiger, J. H. (1980). Statistically based tests for the number of common factors. Paper presented at the Paper presented at the Annual Meeting of the Psychometric Society, Iowa Cyty, 1980.

21. Tanaka, Y., Watadani, S., & Ho Moon, S. (2007). Influence in covariance structure analysis: With an application to confirmatory factor analysis. Communications in Statistics-Theory and Methods, 20(12), 3805-3821. https://doi.org/10.1080/03610929108830742

22. van Kesteren, E.-J., & Oberski, D. L. (2019). Structural equation models as computation graphs. arXiv preprint arXiv:1905.04492.

23. Yuan, K. H., & Bentler, P. M. (2002). Structural equation modeling with robust covariances. Sociological methodology, 28(1), 363-396. https://doi.org/10.1111/0081-1750.00052

24. Yuan, K. H., Fung, W. K., & Reise, S. P. (2010). Three Mahalanobis distances and their role in assessing unidimensionality. British Journal of Mathematical and Statistical Psychology, 57(1), 151-165. https://doi.org/10.1348/000711004849231

25. Yuan, K. H., Tong, X., & Zhang, Z. (2014). Bias and efficiency for SEM with missing data and auxiliary variables: Two-stage robust method versus two-stage ML. Structural Equation Modeling: A Multidisciplinary Journal, 22(2), 178-192. https://doi.org/10.1080/10705511.2014.935750

26. Yuan, K. H., & Zhang, Z. (2012). Robust structural equation modeling with missing data and auxiliary variables. Psychometrika, 77(4), 803-826. https://doi.org/10.1007/s11336-012-9282-4

27. Yuan, K. H., & Zhong, X. (2008). 8. outliers, leverage observations, and influential cases in factor analysis: Using robust procedures to minimize their effect. Sociological methodology, 38(1), 329-368. https://doi.org/10.1111/j.1467-9531.2008.00198.x

28. Yuan, K. H., & Zhong, X. (2013). Robustness of fit indices to outliers and leverage observations in structural equation modeling. Psychological methods, 18(2), 121. https://doi.org/10.1037/a0031604

29. Zhong, X., & Yuan, K.-H. (2011). Bias and efficiency in structural equation modeling: Maximum likelihood versus robust methods. Multivariate Behavioral Research, 46(2), 229-265. https://doi.org/10.1080/00273171.2011.558736