Pemodelan Faktor-Faktor yang Mempengaruhi Jumlah Kasus Diabetes Melitus di Jawa Timur Menggunakan Geographically Weighted Generalized Poisson Regression dan Geographically Weighted Negative Binomial Regression

Elvira Dian Safire (Institut Teknologi Sepuluh Nopember)
Purhadi Purhadi

Inferensi
Vol 6, No 1 (2023)

Diabetes mellitus is a chronic disease of metabolic disorders characterized by blood-sugar levels exceeding normal limits. The province contributing the largest number of cases of diabetes mellitus in Indonesia in 2019 is East Java Province. To know the factors that affect the number of cases of diabetes mellitus is used approach with Geographically Weighted Generalized Poisson Regression (GWGPR) and graphically Weighted Negative Binomial Regression (GWNBR) methods. The highest number of people with diabetes mellitus in East Java is in Surabaya with 94076 cases and the lowest is in Batu City which is 3344 cases. GWGPR and GWNBR modeling both resulted in 4 groups for significants variables in each district/city. The AICc value comparison of the GWGPR and GWNBR models shows almost the same value. Sehingga menunjukkan bahwa model GWGPR dan GWNBR sudah sesuai. So, it shows that the GWGPR and GWNBR models are appropriate. The GWGPR model has a smaller AICc value than the GWNBR model, so the GWGPR method is best suited to model the number of cases of diabetes mellitus in districts/cities in East Java compared to GWNBR method.