PEMODELAN GEOGRAPHICALLY WEIGHTED REGRESSION DAN MIXED GEOGRAPHICALLY WEIGHTED REGRESSION DENGAN FUNGSI PEMBOBOT GAUSSIAN KERNEL (STUDI KASUS: FAKTOR-FAKTOR YANG MEMENGARUHI INDEKS PEMBANGUNAN MANUSIA PADA SETIAP KABUPATEN/KOTA PROVINSI JAWA BARAT TAHUN 2021)

Silvy Mumayya, NIM.: 19106010039 (2023) PEMODELAN GEOGRAPHICALLY WEIGHTED REGRESSION DAN MIXED GEOGRAPHICALLY WEIGHTED REGRESSION DENGAN FUNGSI PEMBOBOT GAUSSIAN KERNEL (STUDI KASUS: FAKTOR-FAKTOR YANG MEMENGARUHI INDEKS PEMBANGUNAN MANUSIA PADA SETIAP KABUPATEN/KOTA PROVINSI JAWA BARAT TAHUN 2021). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (PEMODELAN GEOGRAPHICALLY WEIGHTED REGRESSION (GWR) DAN MIXED GEOGRAPHICALLY WEIGHTED REGRESSION (MGWR) DENGAN FUNGSI PEMBOBOT GAUSSIAN KERNEL) 19106010039_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (3MB) | Preview
	Text (PEMODELAN GEOGRAPHICALLY WEIGHTED REGRESSION (GWR) DAN MIXED GEOGRAPHICALLY WEIGHTED REGRESSION (MGWR) DENGAN FUNGSI PEMBOBOT GAUSSIAN KERNEL) 19106010039_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (6MB) | Request a copy

Abstract

The Geographically Weighted Regression (GWR) model is a development of the global regression model to address the problem of spatial heterogeneity. The resulting parameter estimates are local at each observation location. In testing the parameters of the GWR model, variables are sometimes found that have no effect locally, causing the development of the GWR model to become a Mixed Geographically Weighted Regression (MGWR) model. The MGWR model is a combined modeling of global regression and GWR with the results of parameter estimates that are local and some are global at each observation location. The estimation of GWR and MGWR model parameters uses the Weighted Least Square (WLS) method by assigning weights to each observation location. The spatial weights used are fixed gaussian and adaptive gaussian kernels, selecting the optimum bandwidth using the Cross Validation (CV) method and measuring the goodness of the model using the Akaike Information Criterion (AIC). This study discusses GWR and MGWR modeling on human development index data in West Java Province in 2021. Based on the GWR modeling testing process that is better than the MGWR model, the GWR model with adaptive gaussian kernel weights has the most optimum AIC value of 8.870221. The factors that influence HDI vary locally in each observation including the number of schools, the rate of population growth, the percentage of poor people, the open unemployment rate and the percentage of gross regional domestic product growth.

Share this knowledge with your friends :

Actions (login required)

View Item