Pemodelan Robust Geographically Weighted Regression (RGWR) Penduga-M untuk Analisis Tingkat Keparahan Kemiskinan di Provinsi Jawa Barat Tahun 2023
Abstract
Poverty is influenced by diverse regional conditions, requiring an approach that captures spatial variation. Robust Geographically Weighted Regression with M-estimation (RGWR-M) and Tukey Bisquare weighting was applied to analyze the poverty severity index in districts and cities of West Java Province in 2023. This method was chosen because Ordinary Least Squares (OLS) is prone to heterogeneity and Geographically Weighted Regression (GWR) is sensitive to outliers. Independent variables include the percentage of poor population, access to safe drinking water, population density, and per capita expenditure. Results show that RGWR-M achieved the best performance with a Mean Absolute Deviation (MAD) of 0.0443, and all variables had significant effects. These findings highlight the importance of robust models in spatial analysis to provide accurate estimates and support targeted poverty alleviation policies.
Keywords: Poverty, Robust regression, RGWR-M, Spasial heterogeneity, Tukey bisquare
Kemiskinan dipengaruhi oleh kondisi yang berbeda di setiap wilayah sehingga memerlukan pendekatan yang mampu menangkap variasi spasial. Robust Geographically Weighted Regression dengan pendugaan-M (RGWR-M) dan pembobot Tukey Bisquare digunakan untuk menganalisis indeks keparahan kemiskinan di kabupaten/kota Provinsi Jawa Barat tahun 2023. Metode ini dipilih karena OLS rentan heterogenitas dan GWR sensitif terhadap outlier. Variabel independen yang diuji adalah persentase penduduk miskin, akses air minum layak, kepadatan penduduk, dan pengeluaran per kapita. Hasil penelitian menunjukkan RGWR-M memberikan kinerja terbaik dengan MAD 0,0443, dan semua variabel berpengaruh signifikan. Temuan ini menegaskan pentingnya model robust dalam perumusan kebijakan pengentasan kemiskinan.
Keywords
Full Text:
PDFReferences
Aghayari, M., Pahlavani, P., & Bigdeli, B. (2017). A geographic weighted regression for rural highways crashes modelling using the Gaussian and Tricube kernels: A case study of USA rural highways. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 42, 305-309.
Bivand, R., & Brunstad, R. (2005). Further explorations of interactions between agricultural policy and regional growth in western Europe: Approaches to nonstationarity in spatial econometrics. 45th Congress of the European Regional Science Association: “Land Use and Water Management in a Sustainable Network Society.”, 671-693.
Guo, L., Ma, Z., & Zhang, L. (2008). Comparison of bandwidth selection in application of geographically weighted regression: a case study. Canadian Journal of Forest Research, 38(9), 2526-2534.
Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41(3), 924-936.
Hidayah, N. R. (2024). Analisis dampak pendidikan, perumahan dan pengangguran terhadap tingkat kemiskinan di kabupaten/kota se-Jawa Timur. Jurnal Ilmiah Manajemen, Ekonomi, & Akuntansi (MEA), 8(1), 2095-2104.
Isnaini, B., Syafitri, U. D., & Aidi, M. N. (2019). Estimating the parameters of a robust geographically weighted regression model in gross regional domestics’ product in East Java. International Journal of Sciences: Basic and Applied Research (IJSBAR), 48(3), 150-160.
Pamungkas, R. A., Yasin, H., & Rahmawati, R. (2016). Perbandingan model GWR dengan fixed dan adaptive bandwidth untuk persentase penduduk miskin di Jawa Tengah. Jurnal Gaussian, 5(3), 535-544.
Rahmawati, R., Safitri, D., & Fairuzdhiya, O. U. (2015). Analisis spasial pengaruh tingkat pengangguran terhadap kemiskinan di Indonesia (studi kasus Provinsi Jawa Tengah). Media Statistika, 8(1), 23-30.
Santoso, K. N., Abiyyi, F., Roy, A., & Marselino, K. (2022). Analisis spasial kemiskinan pada masa pemulihan pandemi Covid-19 di Jawa Barat tahun 2021. Jurnal Statistika Dan Aplikasinya, 6(2).
Shapiro, S. S., & Wilk, M. B. (1965). Biometrika trust an analysis of variance test for normality (Complete Samples). Source: Biometrika, 52(34), 591–611.
Yu, H. (2025). Generalized geographically and temporally weighted regression. Computers, Environment and Urban Systems, 117, 102244.
DOI: https://doi.org/10.17509/jem.v13i2.90368
Refbacks
- There are currently no refbacks.
Copyright (c) 2025 Mathematics Study Program, Universitas Pendidikan Indonesia

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.