Application of Robust Regression with M-Estimation in the Analysis of the Human Development Index in Papua Province

Authors

  • Anggrainy Togi Marito Siregar Universitas Cenderawasih
  • Happy Alyzhya Haay Universitas Cenderawasih
  • Dina Jumiatul Fitri Universitas Cenderawasih
  • Rasi Kasim Samosir Universitas Cenderawasih
  • Septian Pioner Boumer Anes Siregar Politeknik Mandiri Bina Prestasi Medan

Keywords:

Human Development Index, Robust Regression, M-Estimation, Huber, Outliers, Papua Province

Abstract

The Human Development Index (HDI) is an important indicator for measuring human development achievements in terms of health, education, and a decent standard of living. In Papua Province, substantial differences across regions may produce extreme observations that can affect the stability of Ordinary Least Squares (OLS) estimates. This study examines the statistical relationships between the HDI and four of its component indicators in Papua Province, identifies influential observations, and applies robust regression with M-estimation (Huber function) as an alternative to OLS. Secondary data from Statistics of Papua Province covered nine regencies/cities over 2021–2025 (45 regency-year observations), analyzed as a pooled regression. The dependent variable was the HDI, and the independent variables were Life Expectancy, Mean Years of Schooling, Expected Years of Schooling, and Per Capita Expenditure. The analysis comprised OLS estimation, classical assumption tests, Weighted Least Squares (WLS) to address heteroscedasticity, and Huber M-estimation using the Iteratively Reweighted Least Squares (IRLS) algorithm with the WLS weights as prior weights. The OLS residuals violated the assumptions of normality and homoscedasticity. Studentized residuals flagged two observations (Mamberamo Raya, 2021–2022) as residual outliers, neither of which was a high-leverage point, and ten observations received Huber weights below 1. In the Huber model, Life Expectancy, Expected Years of Schooling, and Per Capita Expenditure were statistically significant at the 5% level, whereas Mean Years of Schooling was significant only at the 10% level (p = 0.0994), which is weaker evidence. Because the predictors are components of the HDI, the results are interpreted as statistical associations rather than causal effects. The robust regression results indicated that Life Expectancy, Expected Years of Schooling, and Per Capita Expenditure significantly affected HDI at the 5% significance level, while Mean Years of Schooling was significant at the 10% level. Per Capita Expenditure and Expected Years of Schooling were the most statistically significant factors. The Huber robust regression model produced a Residual Standard Error (RSE) of 0.9373 and effectively reduced the influence of outliers, resulting in more stable and representative parameter estimates of human development conditions in Papua Province.

References

Badan Pusat Statistik, Indeks Pembangunan Manusia 2024. Jakarta, Indonesia: Badan Pusat Statistik, 2024.

United Nations Development Programme, Human Development Report 2023/2024: Breaking the Gridlock. New York, NY, USA: UNDP, 2024.

S. Syafri, F. Firdayeti, dan K. Handayani, “Determinants of Human Development Index: Case study of provinces in Indonesia,” dalam Proceedings of the First Lekantara Annual Conference on Public Administration, Literature, Social Sciences, Humanities, and Education, 2022, doi: 10.4108/eai.3-8-2021.2315091.

R. Fadhil, A. M. Soleh, dan Y. Angraini, “Penentuan peubah yang memengaruhi Indeks Pembangunan Manusia di Provinsi Riau tahun 2015–2021 dengan regresi data panel,” Xplore: Journal of Statistics, vol. 12, no. 2, 2023, doi: 10.29244/xplore.v12i2.1139.

D. Q. F. de Menezes, D. M. Prata, A. R. Secchi, dan J. C. Pinto, “A review on robust M-estimators for regression analysis,” Computers & Chemical Engineering, vol. 147, art. no. 107254, 2021, doi: 10.1016/j.compchemeng.2021.107254.

A. Björklund, A. Henelius, E. Oikarinen, K. Kallonen, dan K. Puolamäki, “Robust regression via error tolerance,” Data Mining and Knowledge Discovery, vol. 36, hlm. 781–810, 2022, doi: 10.1007/s10618-022-00819-2.

A. Damayanti dan M. Susanti, “Analisis regresi robust estimasi-M pembobot Huber dan Tukey Bisquare pada tingkat kemiskinan Indonesia,” Jurnal Kajian dan Terapan Matematika, vol. 10, no. 2, 2024, doi: 10.21831/jktm.v10i2.19812.

S. A. Sakinah, P. R. Arum, dan T. W. Utami, “Perbandingan regresi robust estimasi M dengan pembobot Andrew, Tukey Bisquare dan Huber pada data realisasi penanaman modal asing tahun 2023,” UNIMUS Web Conferences, 2025, doi: 10.26714/uwc.v8.391-402.2025.

S. H. Nisa dan N. S. B. Maria, “Determinan Indeks Pembangunan Manusia Provinsi DKI Jakarta tahun 2010–2021,” Diponegoro Journal of Economics, vol. 11, no. 4, hlm. 47–58, 2022, doi: 10.14710/djoe.35412.

D. P. P. Tyas dan N. M. Sukartini, “Determinants of the Human Development Index (HDI) in Indonesia, 2014–2021,” Media Trend, vol. 17, no. 2, 2022, doi: 10.21107/mediatrend.v17i2.14610.

L. E. Nainggolan, D. Lie, N. T. Nainggolan, dan R. T. Siregar, “How determinants of the Human Development Index impact Indonesia,” Indonesian Interdisciplinary Journal of Sharia Economics, vol. 5, no. 2, hlm. 687–700, 2022, doi: 10.31538/iijse.v5i2.2477.

M. A. I. Caniago dan M. G. Wibowo, “Determinants of Human Development Index in Indonesia with Maqashid Sharia approach,” Jurnal Ilmiah Ekonomi Islam, vol. 10, no. 1, hlm. 200–209, 2024, doi: 10.29040/jiei.v10i1.12353.

S. A. Putri, Zilrahmi, D. Permana, dan D. Fitria, “Factors evaluation of Indonesia Human Index Development in 2023 using PLS-SEM method,” UNP Journal of Statistics and Data Science, vol. 2, no. 3, hlm. 373–380, 2024, doi: 10.24036/ujsds/vol2-iss3/214.

J. W. Nugroho et al., “Determinants of Eastern Indonesia’s Human Development Index,” Journal of Economic Education and Entrepreneurship Studies, vol. 5, no. 2, hlm. 254–261, 2024, doi: 10.62794/je3s.v5i2.2457.

S. Syaidatussalihah, F. Mala, R. A. Permata, dan A. A. Khalil, “Analisis faktor-faktor yang mempengaruhi Indeks Pembangunan Manusia di Indonesia tahun 2023,” Jurnal Sains Natural, vol. 3, no. 1, hlm. 20–26, 2025, doi: 10.35746/jsn.v3i1.699.

T. Tsukurimichi, Y. Inatsu, V. N. L. Duy, et al., “Conditional selective inference for robust regression and outlier detection using piecewise-linear homotopy continuation,” Annals of the Institute of Statistical Mathematics, vol. 74, hlm. 1197–1228, 2022, doi: 10.1007/s10463-022-00846-2.

S. Manzoor dan S. Hussain, “A new efficient redescending M-estimator for robust fitting of linear regression models in the presence of outliers,” Mathematical Problems in Engineering, 2021, doi: 10.1155/2021/3090537.

M. N. A. Singgih dan A. Fauzan, “Comparison of M estimation, S estimation, with MM estimation to get the best estimation of robust regression in criminal cases in Indonesia,” Jurnal Matematika, Statistika dan Komputasi, vol. 18, no. 2, hlm. 251–260, 2022, doi: 10.20956/j.v18i2.18630.

“Modified robust ridge M-estimators to improve linear regression performance under multicollinearity and outliers,” Journal of Statistical Computation and Simulation, 2025, doi: 10.1080/00949655.2025.2573859.

P. J. Huber, “Robust estimation of a location parameter,” The Annals of Mathematical Statistics, vol. 35, no. 1, hlm. 73–101, 1964, doi: 10.1214/aoms/1177703732.

P. J. Huber, Robust Statistics. New York, NY, USA: John Wiley & Sons, 1981.

P. J. Rousseeuw dan A. M. Leroy, Robust Regression and Outlier Detection. New York, NY, USA: John Wiley & Sons, 1987.

V. J. Yohai, “High breakdown-point and high efficiency robust estimates for regression,” The Annals of Statistics, vol. 15, no. 2, hlm. 642–656, 1987.

D. A. Belsley, E. Kuh, dan R. E. Welsch, Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. New York, NY, USA: John Wiley & Sons, 1980.

R. D. Cook, “Detection of influential observation in linear regression,” Technometrics, vol. 19, no. 1, hlm. 15–18, 1977.

S. S. Shapiro dan M. B. Wilk, “An analysis of variance test for normality (complete samples),” Biometrika, vol. 52, no. 3–4, hlm. 591–611, 1965.

T. S. Breusch dan A. R. Pagan, “A simple test for heteroscedasticity and random coefficient variation,” Econometrica, vol. 47, no. 5, hlm. 1287–1294, 1979.

R. Koenker, “A note on studentizing a test for heteroscedasticity,” Journal of Econometrics, vol. 17, no. 1, hlm. 107–112, 1981.

D. C. Montgomery, E. A. Peck, dan G. G. Vining, Introduction to Linear Regression Analysis, 5th ed. Hoboken, NJ, USA: John Wiley & Sons, 2012.

H. Glejser, “A new test for heteroskedasticity,” Journal of the American Statistical Association, vol. 64, no. 325, hlm. 316–323, 1969.

W. N. Venables dan B. D. Ripley, Modern Applied Statistics with S, 4th ed. New York, NY, USA: Springer, 2002.

Downloads

Published

2026-11-17

How to Cite

Siregar, A. T. M., Haay, H. A., Fitri, D. J., Samosir, R. K., & Siregar, S. P. B. A. (2026). Application of Robust Regression with M-Estimation in the Analysis of the Human Development Index in Papua Province. Jurnal Ilmiah Multidisiplin Indonesia (JIM-ID), 5(11), 2643–2655. Retrieved from https://ejournal.seaninstitute.or.id/index.php/esaprom/article/view/9117