ASTRIDYA PARAMITA, 101041029 (2012) PEMODELAN ANGKA KEJADIAN MALARIA DI PROVINSI PAPUA DENGAN PENDEKATAN REGRESI SPASIAL (Studi Analisis Regresi Spasial Pada Data Riskesdas 2007). Thesis thesis, UNIVERSITAS AIRLANGGA.

Preview	Text (ABSTRAK) gdlhub-gdl-s2-2013-paramitaas-23154-8.abstr-t.pdf Download (193kB) | Preview
	Text (FULL TEXT) gdlhub-gdl-s2-2013-paramitaas-23154-22 FULL TEXT.pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Spatial regression is a statistical to analyze spatial relationships and predict the response variable values of the predictor variables values are known with respect to the spatial aspects. The objective of this study was to obtain models of the incidence of malaria with a spatial regression approach. The data utilized was secondary data Riskesdas 2007. Factors studied include the fisherman jobs type, open of Waste Water Disposal Chanel (SPAL) and the cattle sheds at risk of malaria incidence. Unit of observation was the 20 districts / cities in the province of Papua. The technique for analyze data were used the thematic mapping of each variable, regression testing classical Ordinary Least Square (OLS) and spatial regression test. Thematic mapping of all variable results indicated that nearest locations tend to have the same category. These results was supported by test results of Moran's I spatial effects that indicate the spatial dependence. The results of Lagrange Multiplier test showed Spatial Autoregressive Models (SAR) was an appropriate spatial regression models for modeling the incidence of malaria in Papua Province. SAR models was produce R square 41.52%, AIC 132 486 and SC 136 047 better than OLS with R square 61.57%, AIC 129 239 and SC 133 691. Conclusion: spatial regression (SAR) provides a better model than the classical regression (OLS) because spatial regression involve location factor where the incident occurred.

Actions (login required)

View Item