Robust estimation for proportional odds model through monte carlo simulation

Zulkifley Mohamed

QR Code Link :
Type :	article
Subject :	QA Mathematics
ISSN :	2332-2071
Main Author :	Zulkifley Mohamed
Additional Authors :	Nor Afzalina Azmee
Title :	Robust estimation for proportional odds model through monte carlo simulation

Place of Production :	Tanjung Malim
Publisher :	Fakulti Sains dan Matematik
Year of Publication :	2021
Notes :	Mathematics and Statistics
Corporate Name :	Universiti Pendidikan Sultan Idris
Web Link :	Click to view web link
PDF Full Text :	Login required to access this item.

Abstract : Universiti Pendidikan Sultan Idris

Ordinal regression is used to model the ordinal response variable as functions of several explanatory variables. The most commonly used model for ordinal regression is the proportional odds model (POM). The classical technique for estimating the unknown parameters of this model is the maximum likelihood (ML) estimator. However, this method is not suitable for solving problems with extreme observations. A robust regression method is needed to handle the problem of extreme points in the data. This study proposes Huber M-estimator as a robust method to estimate the parameters of the POM with a logistic link function and polytomous explanatory variables. This study assesses ML estimator performance and the robust method proposed through an extensive Monte Carlo simulation study conducted using statistical software, R. Measurement for comparisons are bias, RMSE, and Lipsitzs? goodness of fit test. Various sample sizes, percentages of contamination, and residual standard deviations are considered in the simulation study. Preliminary results show that Huber estimates provide the best results for parameter estimation and overall model fitting. Huber?s estimator has reached a 50% breakdown point for data containing extreme points that are quite far from most points. In addition, the presence of extreme points that have only a distance of two times far from most points has no major impact on ML estimates. This means that the estimates for ML and Huber may yield the same results if the model's residual values are between-2 and 2. This situation may also occur for data with a percentage of contamination below 5%. ?2021 by authors, all rights reserved.

References

F. Zulkifli, Z. Mohamed, N. A. Azmee, R. Z. Abidin. Predicting Students’ Final Exam Grades Using Cumulative Odds Model, Int. J. Psychosoc. Rehabil., Vol. 24, No. 8, 8711–8727, 2020.

F. R. Hampel, E. Ronchetti, P. J. Rousseeuw, W. A. Stahel. Robust Statistics: the Approach Based on Influence Functions, Wiley, 1986.

P. J. Huber. Robust Statistics, J. Wiley & Sons, New York, 1981.

P. J. Huber, E. Ronchetti. Robust Statistics, 2nd ed., J. Wiley & Sons, New York, 2009.

R. A. Maronna, R. D. Martin, V. Yohai. Robust Statistics: Theory and Methods, J. Wiley & Sons, New York, 2006.

M. Iannario, A. C. Monti, D. Piccolo, E. Ronchetti. Robust Inference for Ordinal Response Models, Vol. 11, 3407–3445, 2017.

F. Zulkifli, Z. Mohamed, N. A. Azmee, R. Z. Abidin. Integrating Mirt in Ordinal Regression Modeling to Predict End-of-semester Exam Grades, Int. J. Psychosoc. Rehabil., Vol. 24, No. 8, 8693–8710, 2020.

E. Francis. Effects of Some Coding Techniques on Multicollinearity and Model Statistics, Math. Theory Model., Vol. 8, No. 4, 156–167, 2018.

M. Iannario, A. Clara, D. Piccolo. Robustness Issues for CUB Models, TEST, 2016.

C. Croux, G. Haesbroeck, C. Ruwet. Robust Estimation for Ordinal Regression, J. Stat. Plan. Inference, Vol. 143, No. 9, 1486–1499, 2013.

I. Moustaki, M.P. Victoria-Feser. Bounded-Influence Robust Estimation in Generalized Linear Latent Variable Models, J. Am. Stat. Assoc., Vol. 101, No. 474, 644–653, 2006.

J. H. Albert, S. Chib. Bayesian Analysis of Binary and Polychotomous Response Data, J. Am. Stat. Assoc., Vol. 88, No. 422, 669–679, 1993.

J. Albert, S. Chib. Bayesian Residual Analysis for Binary Response Regression Models, Biometrika, Vol. 82, No. 4, 747, 1995.

S. H. Walker, D. B. Duncan. Estimation of the Probability of an Event as a Function of Several Independent Variables, Biometrika, Vol. 54, 167–179, 1967.

P. McCullagh. Regression Models for Ordinal Data, J. R. Stat. Soc. Ser. B, Vol. 42, 109–142, 1980.

P. H. Franses, R. Paap. Quantitative Models in Marketing Research, Cambridge University Press, 2010.

S. M. Shaharudin, N. Ahmad, N. H. Zainuddin, N. S. Mohamed. Identification of Rainfall Patterns on Hydrological Simulation Using Robust Principal Component Analysis, Indones. J. Electr. Eng. Comput. Sci., Vol. 11, No. 3, 1162–1167, 2018.

P. J. Rousseeuw, A. M. Leroy. Robust Regression and Outlier Detection. Hoboken, John Wiley & Sons, Inc., NJ, USA, 1987.

M. O. A. Abu-Shawiesh, M. Riaz, Q. U. A. Khaliq. MTSD-TCC: A Robust Alternative to Tukey’s Control Chart (TCC) Based on the Modified Trimmed Standard Deviation (MTSD), Mathematics and Statistics, Vol. 8, No. 3, 262–277, 2020. DOI: 10.13189/ms.2020.080304.

W. H. Nugroho, N. W. S. Wardhani, A. A. R. Fernandes, Solimun. Robust Regression Analysis Study for Data With Outliers at Some Significance Levels, Mathematics and Statistics, Vol. 8, No. 4, 373–381, 2020. DOI: 10.13189/ms.2020.080401.

Y. Jiang, Y. G. Wang, L. Fu, X. Wang. Robust Estimation Using Modified Huber’s Functions With New Tails, Technometrics, Vol. 61, No. 1, 111–122, 2019.

Y. Susanti, H. Pratiwi, S. S. H., T. Liana. M Estimation, S Estimation, and MM Estimation in Robust Regression, Int. J. Pure Appl. Math., Vol. 91, No. 3, 349–360, 2014.

D. Dasiou, C. Moyssiadis. The 50% Breakdown Point in Simultaneous M-estimation of Location and Scale, Stat. Pap., Vol. 42, 243–252, 2001.

This material may be protected under Copyright Act which governs the making of photocopies or reproductions of copyrighted materials.
You may use the digitized material for private study, scholarship, or research.

Back to previous page