EMAIL: PASSWORD:
Front Office
UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya


Kampus ITS Sukolilo - Surabaya 60111

Phone : 031-5921733 , 5923623
Fax : 031-5937774
E-mail : libits@its.ac.id
Website : http://library.its.ac.id

Support (Customer Service) :
timit_perpus@its.ac.id




Welcome..guys!

Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT


Moh. Fandika Aqsa


Davi Wahyuni


Tondo Indra Nyata


Anis Wulandari


Ansi Aflacha




ITS » Master Theses » Statistika - S2
Posted by aguss at 14/05/2009 19:53:17  •  5530 Views


PERBANDINGAN ESTIMASI TOTAL POPULASI PENDUDUK BERDASARKAN MODEL PENALIZED SPLINE DENGAN ESTIMASI RASIO

THE COMPARISON BETWEEN THE TOTAL ESTIMATION OF POPULATION BASED ON PENALIZED SPLINE MODEL AND RATIO ESTIMATION

Author :
KALE, MATAMIRA B. 




ABSTRAK

Estimasi total populasi berdasarkan sampel dengan peluang inklusi yang tidak sama dikembangkan dengan berbagai metode baik yang berdasarkan desain samplingnya maupun yang berdasarkan model. Estimasi berdasarkan model regresi parametrik seperti Generalized Regression GR menghasilkan estimator yang lebih efisien dibandingkan dengan estimator Horvitz Thompson HT. Kelemahan estimasi berdasarkan model adalah jika model yang dibentuk tidak sesuai dengan pola data maka estimator yang dihasilkan menjadi tidak efisien. Dengan berkembangnya regresi nonparametrik Penalized Spline dengan fleksibilitas yang tinggi maka dengan metode ini kelemahan estimasi berbasis model dapat direduksi. Tulisan ini bertujuan untuk mengkaji prosedur estimasi total populasi berdasarkan model Penalized Spline. Pembahasan disertai dengan perbandingan hasil estimasi total populasi penduduk 0-4 tahun berdasarkan model Penalized Spline dengan metode estimasi rasio yang digunakan Badan Pusat Statistik BPS dalam estimasi total populasi pada Survei Sosial Ekonomi Nasional SUSENAS. Estimasi total populasi berdasarkan model Penalized Spline diperoleh melalui estimasi rata-rata unit sampel tahap pertama berdasarkan model regresi Penalized Spline linear. Estimasi parameter regresi menggunakan kerangka kerja model linear campuran dimana estimasi parameter efek tetap menggunakan metode maximum likelihood dan prediktor parameter efek random berdasarkan kriteria prediktor linear tak bias terbaik. Estimasi komponen varians menggunakan metode Restricted Maximum Likelohood REML. Berdasarkan kriteria MSE Mean Square Error estimasi berdasarkan model Penalized Spline lebih baik dibandingkan dengan estimasi rasio untuk estimasi total populasi penduduk 0-4 tahun di Kota Kupang.


ABSTRACT

The estimation of total population based on sample with an unequal inclusion probability is developed using various kind of methods either those based on its sampling design or based on the model. The estimation which is based on the parametric regression model such as Generalized Regression GR result in a more efficient estimator compared with Horvitz Thompson HT estimator. The weakness of the estimation based on model is that if the model formed is misspecified the estimator will not be efficient. With the progress of the nonparametric regression of Penalized Spline with high flexibility then with this method the weakness of the model-based estimation can be reduced. This thesis is aimed at analysing the estimation procedure of total estimation of population based on Penalized Spline model. The description is associated with the comparison of result of the total estimation of population for people at 0 4 years of age based on Penalized Spline model with the ratio estimation model used by BPS Statistics Indonesian in estimating the total population during National Social Economic Survey SUSENAS. The total estimation of population based on Penalized Spline model is derived from the mean estimation of primary sampling unit based on the regression model of linear Penalized Spline. The estimation of the regression parameter using the framework of linear mixed model where the estimation of fixed effect parameter uses the maximum likelihood method and random effect parameter predictor which based on the best unbiased linear predictor. The estimation of variance components uses Restricted Maximum Likelohood REML method. Based on MSE Mean Square Error criteria the estimation of population total for people at 04 years of age in Kota Kupang based on Penalized Spline model is better than ratio estimation.



KeywordsModel linear campuran ; Penalized Spline ; REML
 
Subject:  Analisis regresi
Contributor
  1. Drs. I Nyoman Latra, MS
    Sodikin Baidowi, M.Stats.
Date Create: 14/05/2009
Type: Text
Format: pdf
Language: Indonesian
Identifier: ITS-Master-3100008031180
Collection ID: 3100008031180
Call Number: RTSt 519.536 Kal p


Source
Master Theses, Statistics, RTSt 519.536 Kal p, 2008

Coverage
ITS Community Only




[ Download - Summary ]

ITS-Master-3100008031180-4143.pdf




 Similar Document...




! ATTENTION !

To facilitate the activation process, please fill out the member application form correctly and completely

Registration activation of our members will process up to max 24 hours (confirm by email). Please wait patiently

POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan





You are connected from 54.144.7.239
using CCBot/2.0 (http://commoncrawl.org/faq/)



Copyright © ITS Library 2006 - 2016 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan