Detection of spatial-temporal autocorrelation using multivariate moran and lisa method on dengue hemorrhagic fever (DHF) incidence, east java, Indonesia

Spatial-temporal autocorrelation is the correlation of a variable with itself over space and time. The study aim is to detect spatial-temporal autocorrelation using multivariate Moran and Lisa Method on Dengue Hemorrhagic Fever (DHF) incidence rate. This uses monthly data on the DHF incidence rate of 38 districts in East Java from 2002 to 2009. The results show that both multivariate Moran and LISA method yield patterns of spatial-temporal autocorrelation of DHF incidence rate that is different from month to month and from region to region. Positive spatial-temporal autocorrelation patterns on the incidence of DHF are found in the following cities/districs Pacitan, Ponorogo, Trenggalek, Tulungagung, Malang, Lumajang, Banyuwangi, Gresik, Lamongan, Mojokerto, Jombang, Sampang, Pamekasan Sumenep (P value <0.05). Whereas in other regions the incidence of DHF show no significant spatial-temporal autocorrelation.