MODEL SPASIAL AUTOTOREGRESIF POISSON UNTUK MENDETEKSI FAKTOR-FAKTOR YANG BERPENGARUH TERHADAP JUMLAH PENDERITA HIV DI PROVINSI JAWA TIMUR
(1) Program Studi Matematika, Jurusan Matematika, FMIPA, Universitas Negeri Jakarta, Indonesia
Abstract
Dalam upaya menangani jumlah pengidap HIV atau AIDS pada tiap kabupaten di Provinsi Jawa Timur diperlukan pengetahuan tentang faktor-faktor yang mempengaruhinya secara spasial maupun nonspasial. Jumlah pengidap HIV atau AIDS merupakan data cacahan (count data) dan kejadian warga mengidap HIV atau AIDS merupakan kejadian yang jarang terjadi, sehingga dalam penelitian ini menggunakan model Spatial Autoregressive Poisson (SAR Poisson). Penggunaan model SAR Poisson bertujuan untuk menentukan faktor-faktor yang berpengaruh secara spasial maupun nonspasial terhadap jumlah pengidap HIV atau AIDS di Provinsi Jawa Timur. Berdasarkan hasil penelitian diperoleh faktor-faktor yang mempengaruhi jumlah pengidap HIV atau AIDS di Provinsi Jawa Timur yaitu: jumlah warga yang tuna susila, jumlah korban penyalahgunaan NAPZA, jumlah keluarga fakir miskin, dan jumlah wanita rawan sosial ekonomi. Pendugaan parameter dalam penelitian ini menggunakan metode maksimum likelihood. Berdasarkan hasil penelitian ini diperoleh korelasi spasial yang signifikan sebesar yang berarti bahwa jumlah pengidap HIV atau AIDS pada suatu wilayah atau lokasi yang berdekatan akan berpengaruh terhadap jumlah pengidap HIV atau AIDS pada tiap kabupaten di Provinsi Jawa Timur pada lokasi di sekitarnya. Koefisien determinasi diperoleh dari model ini sebesar 0,51.
In efforts to handle the number of people living with HIV or AIDS in each municipality in East Java province are required knowledge about the factors that influence spatially and nonspatially. Number of people living with HIV or AIDS is the counting data and the incidence of residents living with HIV or AIDS is very rare, so in this study using a model Spatial Autoregressive Poisson (Poisson SAR). Use of SAR Poisson models aimed to determine the factors that influence spatially and nonspatially to the number of people living with HIV or AIDS in East Java province. Based on the research results was obtained the factors that influence the number of people living with HIV or AIDS in East Java province are the number of people who do prostitution, the number of victims of drug abuse, the number of destitute families, and the number of women prone to socioeconomic. The estimation of the parameters in this study using maximum likelihood method. Based on the results of this study showed significant spatial correlation of ρ = 0.2, which means that the number of people living with HIV or AIDS in a region or a nearby location will affect the number of people living with HIV or AIDS in each municipality in the province of East Java at locations in the vicinity. The coefficient of determination obtained from this model of 0.51.
Keywords
Full Text:
PDFReferences
Anselin L. 1988. Spatial Economics: Methods and Models. Dordrecht: Academic Publishers.
[BPPK] Badan Penelitian dan Pengembangan Kesehatan. 2010. Riset Kesehatan Dasar (Riskesdas). Jakarta: Kementerian Kesehatan Republik Indonesia.
[BPS] Badan Pusat Statistika. 2010. Provinsi Jawa Timur dalam Angka. Jawa Timur: BPS.
Cameron AC & Trivedi PK. 1998. Regression Analysis of Count Data. New York: Cambridge University.
Cameron AC & Windmeijer FAG. 1995. R-squared Measures for Count Data Regession Models with Applications to Health Care Utilization. Journal of Business and Economics Statistics (1995).
Fleiss JL, Levin B & Paik MC. 2003. Statistical Methods for Rates and Proportions. Ed ke-3. USA: Columbia University.
Fotheringham AS & Rogerson PA. 2009. Handbook of Spatial analysis. London: Sage Publications Ltd.
Griffith DA & Haining R. 2006. Beyond Mule Kicks: The Poisson Distribution in Geographical Analysis. Geographical Analysis 38 : 123–139.
Lambert DM, Brown JP & Florax RJGM. 2010. A Two-Step Estimator for a Spatial Lag Model of Counts: Theory, Small Sample Performance and application. USA: Dept. of Agricultural Economics Purdue University.
Lichstein JW, SimonsTR, Shriner SA & Franzreb KE. 2002. Spatial Autocorrelation and Autoregressive Models in Ecology. Ecological Monographs 72:445–463. [http://dx.doi.org/10.1890/0012-9615(2002)072[0445:SAAAMI]2.0.CO;2]
McCulloch CE & Searle SR. 2001. Generalized Linear and Mixed Models. Canada: John Wiley & Sons, Inc.
Taddy MA. 2010. Autoregressive Mixture Models for Dynamic Spatial Poisson Processes: Application to Tracking Intensity of Violent Crime. Journal of the American Statistical Association 105 (492): 1403-1417 [DOI:10.1198/jasa.2010.ap09655]
Wang Y & Kockelman KM. 2013. A Poisson-lognormal conditional-autoregressive model for multivariate spatial analysis of pedestrian crash counts across neighborhoods. Accident Analysis & Prevention 60: 71–84 [doi:10.1016/j.aap.2013.07.030]
Refbacks
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution 4.0 International License.