Pemodelan Faktor-Faktor Kemiskinan di Indonesia Menggunakan Pendekatan Regresi LASSO

widya febriani

doi:10.26877/imajiner.v8i3.27085

Pemodelan Faktor-Faktor Kemiskinan di Indonesia Menggunakan Pendekatan Regresi LASSO

widya febriani

Abstract

In September 2024, the percentage of poor people in Indonesia reached 8.57 percent. This figure shows a decrease of 0.46 percentage points compared to March 2024. However, this poverty rate is still far from the national target. This study aims to analyze the factors that influence the percentage of poor people (P₀) in order to provide a reference for the government to optimize policies on the dominant factors that have a strong impact on the percentage of poor people (P₀). The research method uses LASSO regression due to the high multicollinearity problem in the data used. The analysis results show that the poverty depth variable (P₁) and poverty severity variable (P₂) have a dominant influence on the percentage of poor people (P₀). The coefficient of determination value of 95% on the training data and 96% on the testing data indicates that the model has a very good ability to explain the percentage of poor people (P₀).

Keywords

Poverty, Lasso Regression; Multicollinearity; Poverty; Poverty Depth; Poverty Severity, Indonesia

Full Text:

PDF

References

Alvia, D.O., Efendi, I.W., Malihah, K. P., Kurniawan, M., Islam, U., Raden, N., Lampung, I., & Lampung, K.B (2024). Pengaruh Indeks Pembangunan Manusia dan Pengangguran Terhadap Tingkat Kemiskinan di Indonesia Tahun 2014-2023. Jurnal Media Akademik, 2(6)

Badan Pusat Statistik. (2025). Profil Kemiskinan di Indonesia September 2024 (Badan Pusat Statistik) ; Issue 07). Badan Pusat Statistik.

Celindoni, M (2015). Decomposing Vulnerability To Poverty. Review of Income and Wealth, 1, 59-74.

Chellai, F. (1996). Lasso Regression Model. 0-2.

Darnius, O., Greace, Y., Turnip, C., Tarigan, E.D., & Marpaung, T.J. (2024). Mathematical Modelling of Engineering Problems Poverty Modeling in North Sumatera Province Considering County Location Using Geographical Weighted Regression and LASSO.11(8), 2100-2108

Guha, S.(2024). Title: Feature Selection Using Lasso Regression Enhances Deep Learning Model Performance For Diagnosis OF Lung Cancer from Transcriptomic Data 1 Centre Central University), New Delhi-110025, India Feature Selection Using Lasso Regression Enhances Deep. BioRxiv, 1-20.

Mahalani, A. J., Azizah, N., Rifai, K., Statistika, P., Matematika, F., & Alam, P.(2022). Least Absolute Shrinkage And Selection Operator (LASSO) untuk Mengatasi Multikolinearitas pada Model Regresi Linear Berganda. Bnadung Conference Series: Statistick

DOI: https://doi.org/10.26877/imajiner.v8i3.27085