Amalia, Suci (2024) Model Probit Bivariat Semiparametrik Menggunakan Pendekatan Fisher Scoring (Studi Kasus: Inisiasi Menyusu Dini (IMD) dan Pemberian ASI Eksklusif di Provinsi Jawa Timur Tahun 2021). Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6003212020-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6003212020-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (2MB) | Request a copy |
Abstract
Model probit dengan dua variabel respon disebut model probit bivariat. Model probit bivariat semiparametrik merupakan pengembangan model probit bivariat yang memuat komponen parametrik dan nonparametrik. Komponen nonparametrik berupa fungsi smooth (penghalus) kovariat kontinu. Model probit bivariat semiparametrik mampu menangani masalah nonlinieritas dari variabel prediktor kontinu yang tidak terdeteksi. Dengan demikian, tujuan penelitian ini adalah mengkaji estimasi parameter pada model probit bivariat semiparametrik dan menerapkan model tersebut pada data inisiasi menyusu dini (IMD) dan pemberian ASI eksklusif. Data IMD dan pemberian ASI eksklusif merupakan data sekunder yang diperoleh dari SUSENAS 2021 Jawa Timur sebanyak 2.796 individu. Variabel prediktor yang diduga mempengaruhi IMD dan pemberian ASI eksklusif yakni usia ibu, usia pertama perkawinan, tingkat pendidikan ibu, status pekerjaan ibu dan penolong kelahiran. Kajian estimasi parameter model probit bivariat semiparametrik menggunakan penalized maximum likelihood estimation (PMLE), dalam prosesnya menghasilkan persamaan yang tidak closed form sehingga dibutuhkan iterasi. Iterasi Fisher Scoring dipilih karena memiliki performa yang lebih baik dalam mengatasi adanya kemungkinan non kovergenitas. Pemodelan menggunakan model probit bivariat semiparametrik pada data IMD dan pemberian ASI eksklusif menghasilkan variabel respon IMD signifikan dipengaruhi oleh variabel penolong kelahiran sedangkan variabel respon pemberian ASI eksklusif signifikan dipengaruhi oleh variabel usia ibu, tingkat pendidikan ibu dan penolong kelahiran.
=============================================================================================================================
A probit model with two response variables is called a bivariate probit model. The semiparametric bivariate probit model is a development of the bivariate probit model that contains parametric and nonparametric components. The non-parametric component is a continuous covariate smooth function. The semi-parametric bivariate probit model is able to handle the problem of nonlinearity of undetected continuous predictor variables. Thus, the purpose of this study is to examine parameter estimation in semiparametric bivariate probit models and apply these models to early breastfeeding initiation and exclusive breastfeeding data. early breastfeeding initiation data and exclusive breastfeeding are secondary data obtained from SUSENAS 2021 East Java as many as 2.796 individuals The predictor variables thought to affect early breastfeeding initiation and exclusive breastfeeding were maternal age, first age of marriage, mother's education level, mother's employment status and birth attendant. The study of parameter estimation of semiparametric bivariate probit models using penalized maximum likelihood estimation (PMLE), in the process produces equations that are not closed form so that iteration is needed. The Fisher Scoring iteration was chosen because it has a better performance in overcoming the possibility of non-convergence. Modeling using a semiparametric bivariate probit model on early breastfeeding initiation data and exclusive breastfeeding resulted in a significant early breastfeeding initiation response variable influenced by birth attendant variables while a significant exclusive breastfeeding response variable influenced by maternal age variables, maternal education level and birth attendants.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Fisher Scoring, Inisiasi Menyusui Dini, Pemberian ASI Eksklusif, Penalized Maximum Likelihood Estimation, Probit Bivariat Semiparamterik, Early Breastfeeding Initiation, Exclusive Breastfeeding, Semiparametric Bivariate Probit |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA401 Mathematical models. |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | Suci Amalia |
| Date Deposited: | 18 Feb 2024 12:51 |
| Last Modified: | 18 Feb 2024 12:51 |
| URI: | http://repository.its.ac.id/id/eprint/107333 |
Actions (login required)
 |
View Item |