Penaksiran Parameter dan Statistik Uji Model Mixed Geographically Weighted Bivariate Poisson Inverse Gaussian Regression (Studi Kasus: Jumlah Kematian Bayi dan Jumalh Kematian Ibu di Kota Surabaya Tahun 2015

Arniva, Nendy Septi (2018) Penaksiran Parameter dan Statistik Uji Model Mixed Geographically Weighted Bivariate Poisson Inverse Gaussian Regression (Studi Kasus: Jumlah Kematian Bayi dan Jumalh Kematian Ibu di Kota Surabaya Tahun 2015. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211650010029-Master_Thesis.pdf]
Preview
Text
06211650010029-Master_Thesis.pdf - Accepted Version

Download (3MB) | Preview

Abstract

Regresi poisson merupakan analisis regresi nonlinear dari distribusi poisson yang digunakan dalam menganalisis data diskrit. Pada regresi poisson mensyaratkan kondisi dimana nilai mean dan varians dari variabel respon bernilai sama atau kondisi equidispersion. Namun dalam kasus banyak terjadi overdispersion atau underdispersion. Mixed poisson distribution merupakan solusi alternatif untuk kasus overdispersi maupun underdispersi. Salah satu metode untuk mengatasinya adalah distribusi Poisson Inverse Gaussian (PIG). Pada Poisson Inverse Gaussian tidak semua kasus yang hanya melibatkan satu varibel respon, karena dalam kenyataannya beberapa kasus akan melibatkan lebih dari satu variabel respon. Dalam penelitian ini dilakukan pengembangan model regresi bivariat yang melibatkan faktor spasial yaitu dengan adanya pembobot geografis. Pada kenyataannya tidak semua variabel dalam model geographically weighted regression mempunyai pengaruh secara spasial, ada beberapa variabel prediktor berpengaruh secara global. Penelitian ini menghasilkan estimator parameter model menggunakan Maximum Likelihood Estimation (MLE) dengan iterasi Newton-Raphson. Selanjutnya mendapatkan statistik uji pada model Mixed Geographically Weighted Bivariate Poisson Inverse Gaussian Regression (MGWBPIGR) menggunakan MLRT. Penerapan model Mixed Geographically Weighted Bivariate Poisson Inverse Gaussian Regression yang terbentuk variabel prediktor yang berpengaruh secara signifikan terhadap jumlah kematian bayi dan jumlah kematian ibu di Kota Surabaya tahun 2015 adalah variabel rasio tenaga kesehatan, persentase persalinan oleh tenaga kesehatan, persentase ibu hamil mendapatkan tablet Fe3, persentase rumah tangga ber-PHBS dan rasio puskesmas.
====================================================================================================
Poisson regression is a nonlinear regression analysis of the poisson distribution used in analyzing discrete data. In poisson regression requires conditions where the mean and variance values of the response variable are equal or equidistpersion conditions. But in case of over dispersion or under dispersion occurs. Mixed poisson distribution is an alternative solution for over dispersion and under dispersion cases. One method to overcome this is the Poisson Inverse Gaussian (PIG) distribution. In Poisson Inverse Gaussian not all cases involve only one response variable, because in reality some cases will involve more than one response variable. This research, the development of bivariate regression model which involves spatial factor that is with the geographic weighting. In fact, not all variables in the geographically weighted regression model have spatial influence, there are several predictor variables globally. The result of this studied is parameter estimation using Maximum Likelihood estimation (MLE) with Newton-Raphson iteration. Furthermore, statistic test on the Mixed Geographically Weighted Bivariate Poisson Inverse Gaussian Regression Model (MGWBPIGR) model. The application of the Mixed Geographically Weighted Bivariate Poisson Inverse Gaussian Regression model which is the predictor variable that significantly affects the number of infant mortality and the number of maternal deaths in Surabaya City 2015 is the ratio of health personnel, the percentage of births by health personnel, the percentage of pregnant women get Fe3 tablets, percentage of households with PHBS and ratio of puskesmas.

Item Type: Thesis (Masters)
Additional Information: RTSt 519.536 Arn p-1
Uncontrolled Keywords: GWR, Kematian Bayi, Kematian Ibu, MGWBPIGR, Poisson Inverse Gaussian, Regresi Bivariate
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Q Science > QA Mathematics > QA9.58 Algorithms
Divisions: Faculty of Mathematics, Computation, and Data Science > Statistics > 49101-(S2) Master Thesis
Depositing User: Arniva Nendy Septi
Date Deposited: 17 Jun 2021 06:34
Last Modified: 17 Jun 2021 06:34
URI: http://repository.its.ac.id/id/eprint/54278

Actions (login required)

View Item View Item