A Comparison of Weights Matrices on Computation of Dengue Spatial Autocorrelation

2nd International Conference on Mathematics, Science, Education and Technology

数学;自然科学;教育

Suryowati, K.^1 ; Bekti, R.D.^1 ; Faradila, A.^1

Department of Statistics, Institut Sainsand Teknologi AKPRIND Yogyakarta, 55222, Indonesia^1

关键词: Dengue hemorrhagic fever;    Inverse distance weight;    Spatial analysis;    Spatial autocorrelations;    Spatial influences;    Spatial weight matrixes;    Weighting functions;    Weighting matrices;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/335/1/012052/pdf
DOI  :  10.1088/1757-899X/335/1/012052

来源: IOP

【 摘 要 】

Spatial autocorrelation is one of spatial analysis to identify patterns of relationship or correlation between locations. This method is very important to get information on the dispersal patterns characteristic of a region and linkages between locations. In this study, it applied on the incidence of Dengue Hemorrhagic Fever (DHF) in 17 sub districts in Sleman, Daerah Istimewa Yogyakarta Province. The link among location indicated by a spatial weight matrix. It describe the structure of neighbouring and reflects the spatial influence. According to the spatial data, type of weighting matrix can be divided into two types: point type (distance) and the neighbourhood area (contiguity). Selection weighting function is one determinant of the results of the spatial analysis. This study use queen contiguity based on first order neighbour weights, queen contiguity based on second order neighbour weights, and inverse distance weights. Queen contiguity first order and inverse distance weights shows that there is the significance spatial autocorrelation in DHF, but not by queen contiguity second order. Queen contiguity first and second order compute 68 and 86 neighbour list.