A Comparison of Multivariate Adaptive Regression  Spline and Spline Nonparametric Regression on Life Expectancy in Indonesia

Bagas Shata Pratama; Suliyanto Suliyanto; M. Fariz Fadillah Mardianto; Sediono Sediono

doi:10.31764/jtam.v9i3.29413

A Comparison of Multivariate Adaptive Regression Spline and Spline Nonparametric Regression on Life Expectancy in Indonesia

Bagas Shata Pratama, Suliyanto Suliyanto, M. Fariz Fadillah Mardianto, Sediono Sediono

Abstract

Life expectancy is a key indicator of a population’s overall health and well-being. It also reflects the effectiveness of government efforts in improving public welfare. Despite various initiatives by both the government and society to improve life expectancy in Indonesia, significant disparities remain. This quantitative study aims to support these efforts by analyzing factors influencing life expectancy in Indonesia using data from the Indonesian Central Agency of Statistics (BPS) in 2023. A comparative analysis was conducted using two methods: Multivariate Adaptive Regression Spline (MARS) and Spline Nonparametric Regression. The results show that the MARS model outperforms the Spline model, achieving a lower Mean Squared Error (MSE) of 1.183 and a higher R-Square of 82.7%. Key variables significantly influencing life expectancy include access to decent housing, access to safe drinking water, per capita expenditure, and the Gini ratio. The findings not only confirm the presence of complex interactions among predictor variables effectively captured by the MARS method, but also contribute to the existing literature by emphasizing the importance of socioeconomic determinants in health outcomes. From a policy perspective, the results suggest that government strategies should prioritize improving access to basic needs and reducing inequality. These insights can guide targeted, data-driven interventions aimed at enhancing life expectancy in Indonesia.

Keywords

Life Expectancy; Regression; Multivariate Adaptive Regression Spline; Spline Nonparametric Regression.

Full Text:

DOWNLOAD [PDF]

References

Ali, P., & Younas, A. (2021). Understanding and interpreting regression analysis. Evidence Based Nursing, 24(4), 116–118. https://doi.org/10.1136/ebnurs-2021-103425

Badan Pusat Statistik. (2023). Angka Harapan Hidup Tahun 2021-2023. BPS Provinsi Sulawesi Barat.

Badan Pusat Statistik. (2024). Umur Harapan Hidup Saat Lahir (UHH). BPS RI. bps.go.id

Bangun, R. H. (2019). Analisis Determinan Angka Harapan Hidup Kabupaten Mandailing Natal(Life Expectations Determinants Analysis In Mandailing Natal Regency). Jurnal Akuntansi Dan Ekonomi, 4(3), 22–31. https://doi.org/10.29407/jae.v4i3.13257

Bekar Adiguzel, M., & Cengiz, M. A. (2023). Model selection in multivariate adaptive regressions splines (MARS) using alternative information criteria. Heliyon, 9(9). https://doi.org/10.1016/j.heliyon.2023.e19964

Erlyn, P., Hidayat, B., Cahyo, A., & Saksono, H. (2022). Investment in Human Resources to Increase Achievement Levels of Sustainable Development. Jurnal Bina Praja, 14(1), 135–146. https://doi.org/10.21787/jbp.14.2022.135-146

Friedman, J. H. (1991). Multivariate Adaptive Regression Splines. The Annals of Statistics, 19(1), 1–141.

Ghozali, I. (2016). Aplikasi Analisis Multivariete Dengan Program IBM SPSS 23. Edisi 8. Badan Penerbit Universitas Diponegoro.

Hasanah, U. (2017). Pengaruh Ketimpangan Pendapatan, Pendapatan per Kapita, dan Pengeluaran Pemerintah di Bidang Kesehatan Terhadap Sektor Kesehatan di Indonesia. Jurnal Ilmu Ekonomi Terapan, 2(1), 30–43. https://doi.org/10.20473/jiet.v2i1.5504

Kuncoro, M. (2019). Metode Riset Untuk Bisnis dan Ekonomi (Edisi ke-3). Erlangga.

Lembang, F. K., Patty, H. W. M., & Maitimu, F. (2019). Analisis Kemiskinan di Kabupaten Maluku Tenggara Barat Menggunakan Pendekatan Multivariate Adaptive Regression Spline (MARS). MEDIA STATISTIKA, 12(2), 188. https://doi.org/10.14710/medstat.12.2.188-199

Maharani, M., & Saputro, D. R. S. (2021). Generalized Cross Validation (GCV) in Smoothing Spline Nonparametric Regression Models. Journal of Physics: Conference Series, 1808(1), 1–6. https://doi.org/10.1088/1742-6596/1808/1/012053

Mahmoud, H. F. F. (2021). Parametric Versus Semi and Nonparametric Regression Models. International Journal of Statistics and Probability, 10(2), 90–108. https://doi.org/10.5539/ijsp.v10n2p90

Mattalunru, M. R., Annas, S., & Aidid, M. K. (2022). Aplikasi Multivariate Adaptive Regression Splines (MARS) untuk Mengetahui Faktor yang Mempengaruhi Curah Hujan di Kota Makassar. VARIANSI: Journal of Statistics and Its Application on Teaching and Research, 4(1), 9–19. https://doi.org/https://doi.org/10.35580/variansiunm2

Mukrom, M. H., Yasin, H., & Hakim, A. R. (2021). Pemodelan Angka Harapan Hidup Provinsi Jawa Tengah menggunakan Robust Spatial Durbin Model. Jurnal Gaussian, 10(1), 44–54. https://doi.org/https://doi.org/10.14710/j.gauss.10.1.44-54

Nurhuda, G. N., Wasono, W., & Nohe, D. A. (2022). Nonparametric Regression Modeling Based on Spline Truncated Estimator on Simulation Data. Jurnal Matematika, Statistika Dan Komputasi, 19(1), 172–182. https://doi.org/10.20956/j.v19i1.21534

Pratama, Y. M., Fernandes, A. A. R., Wardhani, N. W. S., & Hamdan, R. (2024). Nonparametric Smoothing Spline Approach in Examining Investor Interest Factors. JTAM (Jurnal Teori Dan Aplikasi Matematika), 8(2), 425–440. https://doi.org/10.31764/jtam.v8i2.20192

Purnama, D. I. (2020). A Comparison between Nonparametric Approach: Smoothing Spline and B-Spline to Analyze The Total of Train Passangers in Sumatra Island. EKSAKTA: Journal of Sciences and Data Analysis, 1(1), 73–80. https://doi.org/10.20885/EKSAKTA.vol1.iss1.art11

Risambessy, S., Aulele, S. N., & Lembang, F. K. (2022). Misclassification Analysis of Elementary School Accreditation Data in Ambon City Using Multivariate Adaptive Regression Spline. Jurnal Matematika, Statistika Dan Komputasi, 18(3), 394–406. https://doi.org/10.20956/j.v18i3.19451

Ritonga, N. A. R., & Sutarman. (2023). Estimation of Multivariate Adaptive Regression Splines (MARS) Model Parameters by Using Generalized Least Square (GLS) Method. JMEA : Journal of Mathematics Education and Application, 2(2), 62–72. https://doi.org/10.30596/jmea.v2i2.13106

Wicaksono, W., Wilandari, Y., & Suparti. (2014). Pemodelan Multivariate Adaptive Regression Splines (MARS) pada Faktor-Faktor Resiko Angka Kesakitan Diare (Studi Kasus : Angka Kesakitan Diare di Jawa Tengah, Jawa Timur dan Daerah Istimewa Yogyakarta Tahun 2011). Jurnal Gaussian, 3(2), 253–262. https://doi.org/https://doi.org/10.14710/j.gauss.3.2.253%20-%20262

Worldometer. (2024). Life Expectancy of the World Population. https://www.worldometers.info/demographics/life-expectancy/

Yasmirullah, S. D. P., Otok, B. W., Purnomo, J. D. T., & Prastyo, D. D. (2021). Parameter Estimation of Multivariate Adaptive Regression Spline (MARS) with Stepwise Approach to Multi Drug-Resistant Tuberculosis (MDR-TB) Modeling in Lamongan Regency. Journal of Physics: Conference Series, 1752(1), 1–9. https://doi.org/10.1088/1742-6596/1752/1/012017

DOI: https://doi.org/10.31764/jtam.v9i3.29413