Pemodelan Pemberian ASI Eksklusif Dan Status Bekerja Ibu Dengan Regresi Probit Bivariat Rekursif (Studi Kasus Di Kota Surabaya Tahun 2017)

Isnaini, Fadhila (2019) Pemodelan Pemberian ASI Eksklusif Dan Status Bekerja Ibu Dengan Regresi Probit Bivariat Rekursif (Studi Kasus Di Kota Surabaya Tahun 2017). Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 06211750010016-Master_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Metode pemodelan antara dua variabel biner yang saling dependen yang biasa digunakan adalah metode probit bivariat. Namun ada beberapa kasus khusus dimana salah satu variabel dependen menjadi variabel eksogen bagi variabel dependen yang lain. Metode yang dapat digunakan dalam kasus tersebut adalah regresi probit bivariat rekursif. Regresi probit bivariat rekursif merupakan metode dimana dua persamaan probit yang error-error-nya saling berkorelasi, dan salah satu variabel dependen biner menjadi variabel eksogen untuk variabel dependen lainnya. Hal ini biasa disebut dengan masalah endogenitas. Penelitian ini menggunakan kasus pemberian ASI ekslusif dan status ibu bekerja. Diduga variabel status ibu bekerja dapat mempengaruhi pemberian ASI eksklusif sehingga terdapat masalah endogenitas. Estimasi parameter model Regresi Probit Bivariat menggunakan metode MLE. Turunan kedua dari fungsi ln likelihood menghasilkan persamaan yang tidak closed form sehingga digunakan iterasi IRLS. Variabel yang signifikan terhadap ASI Eksklusif adalah usia ibu, pendidikan Ibu, jumlah persalinan, IMD, dan pendidikan Ayah; sedangkan variabel yang signifikan terhadap status bekerja Ibu adalah jumlah persalinan dan banyak anggota keluarga.
================================================================================================================================
The modeling method between two dependent binary variables that commonly used is the bivariate probit method. There are some special cases where one dependent variable becomes the regressor endogenous variable for the other dependent variable. The method that can be used in this case is recursive bivariate probit regression. Recursive bivariate probit regression is a method in which two probit equations whose errors are correlated, and one of the binary dependent variables being the regressor endogenous variable for the other dependent variable. The interpretation of the parameter correlation of the bivariate probit model is different from the recursive bivariate probit parameter correlation. So the recursive bivariate probit does not consider the correlation parameters in the analysis. Correlation of errors between the two response variables that are close to 0 can be interprated that the two variables can be modeled independently of each other. In fact, this small correlation can be an estimation effect of the bivariate probit model whose data has endogenity problem. Exclusive breastfeeding is measured to identify its characteristics so that policies can be formulated to increase the rate of exclusive breastfeeding. The variable of mother working status is endogenous. Therefore, a recursive model is used to model this case. Variables that significant to exclusive breastfeeding are age, mother’s education, number of childbirth, EIB (Early Initiation Breastfeeding), and father’s education; then variables that significant to mother working status is number of childbirth and number of family member.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	ASI Eksklusif, IRLS, Probit Bivariat Rekursif, Status Bekerja Ibu
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. H Social Sciences > HA Statistics > HA31.7 Estimation
Divisions:	Faculty of Mathematics, Computation, and Data Science > Statistics > 49101-(S2) Master Thesis
Depositing User:	Isnaini Fadhila
Date Deposited:	23 Jul 2026 04:22
Last Modified:	23 Jul 2026 04:22
URI:	http://repository.its.ac.id/id/eprint/68612

Actions (login required)

View Item