Pemodelan Mixed Geographically Weighted Regression (MGWR)

Nur Fajar Apriyani (Laboratorium Statistika Terapan FMIPA Universitas Mulawarman)
Desi Yuniarti (Laboratorium Statistika Ekonomi dan Bisnis FMIPA Universitas Mulawarman)
Memi Nor Hayati (Laboratorium Statistika Terapan FMIPA Universitas Mulawarman)

Eksponensial
Vol 9 No 1 (2018)

Diarrhea disease is one of the conditions which a person has soft or liquid defecate consistency, even can be water and frequency more often in one day. The province of East Kalimantan includes areas where the percentage of diarrhea tends to increase annually. Therefore, as one of the efforts to handle cases of diarrhea in East Kalimantan Province, so that the research using Mixed Geographically Weighted Regression (MGWR) model which is a modeling method that combines global regression model and Geographically Weighted Regression (GWR) model. Modeling MGWR aim to find out the factors that affect the number of diarrhea sufferers, where factors are differentiated into factors that affect locally in each District/City and factors that affect globally throughout the District/City. The result of the research using the MGWR method, the variable of the number of households that live clean and healthy and the number of food management places do not meet the criteria affect globally. The number of communal latrine facilities affect locally.