Modeling of Food Insecurity and Poverty with Geographically Weighted Multivariate Linear Model in Kabupaten Sampang

Yusrina Nur Dianati (Program Studi Magister Statistika Jurusan Matematika Fakultas MIPA Universitas Brawijaya Malang)
Ni Wayan Surya Wardhani (Program Studi Statistika Jurusan Matematika Fakultas MIPA Universitas Brawijaya Malang)
Rahma Fitriani (Program Studi Statistika Jurusan Matematika Fakultas MIPA Universitas Brawijaya Malang)

Natural B
Vol 2, No 3 (2014)

The problem of food insecurity has long been the focus of attention and is very closely related to the problem of poverty in which the two are interrelated phenomena that have a causal relationship. Food insecurity and poverty is a package that is always the problem faced by the government both central and local government, especially in Sampang. Spatial regression models that have been described in general a univariate spatial model, in which the observations have only one response variable that depends on the location of the observation. Geographically Weighted Multivariat Linier Model a multivariate regression models were used to spatially resolve the influence of spatial heterogeneity caused by differences in the conditions of the location with another location. The purpose of this study was to establish Geographically Weighted Multivariat Linier Model (GWMLM) with a weighted cross – variogram gaussian on the problem of food insecurity and poverty in Sampang. Food insecurity and poverty is a phenomenon of spatial heterogeneity.  Based on the 10 sampled villages gained influence the percentage of households without access to electricity (X1), the percentage of main roads are adequate (X2) , the number of health facilities (X3) , and the percentage of malnutrition children (X4) against food insecurity and poverty differently in each location.