Geographically weighted regression modeling for analyzing spatial heterogeneity on relationship between dengue hemorrhagic fever incidence and rainfall in Surabaya, Indonesia

Abstract

Geographically weighted regression (GWR) modeling has been extended to evaluate spatial heterogeneity on the relationship between dengue hemorrhagic fever (DHF) incidence and rainfall in Surabaya, Indonesia. We employed monthly data in 2010 as repeated observation for each subdistrict in Surabaya, subdistrict was then considered as spatial unit. Problem of temporally correlated errors in this modeling was solved by means of data transformation. The GWR model was compared with global regression model using some statistical criteria. The GWR model reveals that the relationship between the DHF incidence and the rainfall is significantly varied in every subdistrict. The influence of the rainfall on the DHF incidence was greater over southeastern subdistricts than other subdistricts in Surabaya. This result holds an important consequence on policy making for regulation in preventing DHF infection according to local characteristic climate in a certain region.