Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/52186
Title: Estimation of Small Area Statistics using Beta-Binomial Model
Pendugaan Statistik Area Kecil dengan Menggunakan Model Beta-Binomial
Authors: Wijayanto, Hari
Rahman, La Ode Abdul
Abadi, Slamet
Keywords: Bayes empirical
Beta-Binomial
Kleinman
Jackknife
Bootstrap
Issue Date: 2011
Abstract: Small area estimation is commonly used to describe smaller domain or population. Small Area Estimation is an important technique to estimate parameter of smaller domain borrowing strength of population parameter estimate through statistics models with random effect. This research is focused in modeling binary data in small area estimation with empirical bayes method. This method is applicable more generally in the sense of handling models for binary and count data. The Beta-binomial model can be used to calculate the proportion of each small area and its variance. This model estimates the parameters of proportion using momen Kleinman method of Beta-Binomial model and the we also compare the MSE of indirect estimation using Naïve, Jackknife, and Bootstap methods. The result shaved that the MSE of indirect estimation lower than the direct estimation. Moreover, the MSE of indirect estimation using Naïve, Jackknife, and Bootstap methods relatively the same. This indirect estimation using Beta-Binomial model were applied to analyze the proportion of poor household in Bekasi district. The result showed that Jakasampurna, Ciketingudik, Bintara Jaya, Jatiluhur, Cikiwul, Mustika Jaya, and Perwira and could be categorized as villages having more poor household.
Suatu pendugaan untuk meningkatkan ukuran contoh dan menurunkan galat baku adalah pendugaan tak langsung. Pendugaan ini memanfaatkan informasi tambahan yang diperoleh dari area kecil lain yang memiliki karakteristik yang serupa. Menurut Rao (2003) prosedur pendugaan area kecil pada dasarnya memanfaatkan informasi dari area itu sendiri, area sekitarnya atau bahkan survei yang berbeda. Pendugaan area kecil bermanfaat untuk menduga parameter area yang berukuran contoh kecil. Pada data biner, model Beta-Binomial dapat digunakan untuk menduga parameter area kecil. Ada dua metode dalam pendugaan area kecil untuk data biner, yaitu metode Bayes empirik dan Bayes hirarki. Penelitian ini menggunakan metode Bayes empirik yang mampu menampung informasi antar area dan mereduksi kuadrat tengah galat.
URI: http://repository.ipb.ac.id/handle/123456789/52186
Appears in Collections:MT - Mathematics and Natural Science

Files in This Item:
File Description SizeFormat 
2011sab.pdf
  Restricted Access
Full Text2.56 MBAdobe PDFView/Open
Abstract.pdf
  Restricted Access
Abstrak307.37 kBAdobe PDFView/Open
BAB I Pendahuluan.pdf
  Restricted Access
Bab I312.62 kBAdobe PDFView/Open
BAB II Tinjauan Pustaka.pdf
  Restricted Access
Bab II370.24 kBAdobe PDFView/Open
BAB III Metodologi Penelitian.pdf
  Restricted Access
Bab III360.42 kBAdobe PDFView/Open
BAB IV Hasil dan Pembahasan.pdf
  Restricted Access
Bab IV1.58 MBAdobe PDFView/Open
BAB V Kesimpulan dan Saran.pdf
  Restricted Access
Bab V307.18 kBAdobe PDFView/Open
Cover.pdf
  Restricted Access
Cover331.34 kBAdobe PDFView/Open
Daftar Pustaka.pdf
  Restricted Access
Daftar Pustaka323.72 kBAdobe PDFView/Open
Lampiran.pdf
  Restricted Access
Lampiran856.74 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.