Skip to content

Compared Ordinary Least Square (OLS) and Geographically Weighted Regression (GWR) using R programming with interpretation

Notifications You must be signed in to change notification settings

michstg/Geographically-Weighted-Regression-GWR-Poverty-North-Sumatera

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Geographically Weighted Regression GWR Using R Programming

Dataset = Data Kemiskinan di Provinsi Sumatera Utara tahun 2018

  • Assumption Hypotesis Testing
  • Compared Model OLS - GWR with classical assumption and Exploratory Data Analysis
  • Metrics Evaluation Model is R-Squared and AIC

Independet Variables which is

  • Y = Persentase Kemiskinan (manual calculation)
  • X1 = Jumlah Penduduk (Web BPS)
  • X2 = Tingkat Pengangguran Terbuka
  • X3 = Produk Domestik Regional Bruto
  • X4 = Indeks Pembangunan Manusia
  • X5 = Upah Minimum (Web BPS)
  • lon = Longitude (Google Maps)
  • lat = Latitude (Google Maps)

GWR =

  • Pembobotan Kernel Gaussian
  • Bandwith optimum = 195.1136
  • CV Score = 657.5274
  • Map Visualization (GGMAP PLOT)