Pemodelan Indeks Pembangunan Manusia (IPM) Metode Baru Menurut Provinsi Tahun 2015 Menggunakan Geographically Weighted Regression (GWR)

Akbar Maulana, Renny Meilawati, Vita Widiastuti


The Human Development Index (HDI) is a parameter of quality of life for an area. The HDI explains how residents can access the results of development in obtaining income, health and education. One method that can be used to find out the factors that influence the human development index in modeling is regression analysis of ordinary least square (OLS). In the Human Development Index data, there is a dependency between measuring data and the location of a region. Therefore, spatial regression analysis can be used in this study. The local form of spatial regression analysis is geographically weighted regression (GWR). GWR shows the existence of spatial heterogeneity (location). This study compares between OLS regression and GWR in the new human development index method by province in 2015. In the GWR model we use fixed Gaussian kernel and kernel fixed bisquare as weighted function. The optimal bandwidth value is obtained by minimizing the cross validation (CV) and Akaike information criterion (AIC) coefficients. The results showed that the GWR model with Gaussian kernel function is better than GWR with bisquare kernel function and OLS model.

Keywords: human development index, ordinary least square, geographically weighted regression, kernel fixed Gaussian,  kernel fixed bisquare

Full Text:



United Nations Development Programme. Human Development Report 2000. Oxford Oxford University Press. New York. 2000.

Rencher, A.C. and Schaalje, G.B. Linier Models in Statistics. John Wiley & Sons Inc. Singapore. 2000.

Anselin, L. Spatial Econometrics: Methods and Models. Kluwer Academic Publishers. Dordrecht. 1998.

Badan Pusat Statistik Nasional. Indeks Pembangunan Manusia Metode Baru Tahun 2015. BPS. Jakarta. 2015.

Charlton, M. and Fotheringham, A.S. Geographically Weighted Regression White Paper. National Centre for Geocomputation. Maynooth University. 2009.

Chasco, C., Garcia, I., & Vicens, J. Modeling Spastial Variations in Household Disposible Income with Geographically Weighted Regression. Munich Personal RePEc Arkhive (MPRA). Working Papper No. 1682. 2007.

Fotheringham, A.S., Brunsdon, C., dan Charlton, M. Geographically Weighted Regression: The Analysis of Spatially Varying Relationships. Jhon Wiley & Sons, LTD. 2002.

Nurdim, F.E. Estimasi dan Pengujian Hipotesis Geographically Weighted Regression Studi Kasus Produktivitas Padi Sawah di Jawa Timur. Surabaya: Jurusan Statistika FMIPA ITS. 2008.


  • There are currently no refbacks.