Pemodelan Faktor Risiko Penyakit Campak pada Balita di Provinsi DKI Jakarta: Pemodelan Faktor Risiko Penyakit Campak pada Balita di Provinsi DKI Jakarta

Ayu Annisa Rahmah; Itasia Dina Sulvianti; Cici Suhaeni; Bimandra Adiputra Djaafara

doi:10.29244/xplore.v9i1.158

Authors

Ayu Annisa Rahmah Departement of Statistics, IPB
Itasia Dina Sulvianti Departement of Statistics, IPB
Cici Suhaeni Departement of Statistics, IPB
Bimandra Adiputra Djaafara Eijkman-Oxford Clinical Research Unit

DOI:

https://doi.org/10.29244/xplore.v9i1.158

Keywords:

Gaussian kernel function, geographically weighted regression, incidence rate of measles, spatial heterogeneity

Abstract

Measles is one of the infectious caused by virus. The disease is easily transmitted and has become one of the main causes of child mortality especially toddlers. In 2016, Jakarta experienced the highest measles case in the last ten years and found the difference in the number of measles cases in each sub-district of Jakarta. This can be caused by the existence of effect of spatial location i.e. spatial heterogeneity. Geographically weighted regression (GWR) is a method that can be applied to address the presence of spatial heterogeneity in the process of developing the model. In this study, the weighting function used was the Gaussian kernel. The modelling process generated 42 local models at sub-district level. Explanatory variables that influence the incidence rate of measles in toddlers (Y) signiï¬cantly are the percentage of immunization coverage measles (X1), the total annual rainfall (X4), and the percentage of the number of toddlers (X5). In this study, the GWR model is better than multiple linear regression model which were indicated by higher value of and smaller value of AIC.

Pemodelan Faktor Risiko Penyakit Campak pada Balita di Provinsi DKI Jakarta