Penaksiran Parameter Bivariate Spatial Error Model Dengan Pendekatan Berndt-Hall-Hall Hausman

Alpain, Izzu Syifa'ian (2025) Penaksiran Parameter Bivariate Spatial Error Model Dengan Pendekatan Berndt-Hall-Hall Hausman. Masters thesis, Institut Teknologi Sepuluh Nopember.

	Text Izzu Syifa'ian_Tesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy
	Text TESIS IZZU SYIFA'IAN 4 Feb 2025.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (3MB) | Request a copy

Abstract

Regresi spasial adalah metode regresi yang dapat digunakan pada data dengan mempertimbangkan efek spasial atau lokasi dalam data yang dianalisis. Pemodelan spasial terdapat dua tipe yaitu pemodelan data berdasarkan titik dan area. Berdasarkan pendekatan area, model regresi spasial dapat menggunakan beberapa model, salah satunya adalah Spatial Error Model (SEM). SEM merupakan model spasial error dimana pada error terdapat korelasi spasial. SEM mempertimbangkan bahwa kesalahan dalam model regresi dapat berkorelasi secara spasial, yang berarti bahwa kesalahan pengamatan di satu lokasi dapat mempengaruhi kesalahan di lokasi lain. Model SEM dikembangkan dengan dua variabel respon yang disebut dengan Bivariat Spasial Error Model (BSEM). Pada penelitian ini, metode BSEM diaplikasikan untuk menganalisis hubungan dua variabel yaitu Indeks Pendidikan dan Indeks Pengeluaran Perkapita. Untuk mengevaluasi model yang dihasilkan, penelitian ini menggunakan Akaike Information Criterion (AIC) dan R-Square. Penaksiran parameter dilakukan menggunakan Maximum Likelihood Estimation (MLE). Pengujian secara serentak dilakukan menggunakan MLRT, penaksir parameter model BSEM diperoleh menggunakan MLE dengan pendekatan algoritma Berndt-Hall-Hall-Hausman (BHHH). Hasil penelitian menunjujkan bahwa model BSEM cukup baik untuk memodelkan dan memprediksi persentase kasus indeks pendidikan dan indeks pengeluaran pada kabupaten/kota di Jawa Timur2022. Dengan variabel respon yang berpengaruh terhadap model yaitu persentase penduduk miskin, kepadatan penduduk, dan rasio guru persiswa dan untuk indeks pengeluaran adalah persentase penduduk miskin, dan kepadatan penduduk.
=============================================================================================================================
Spatial regression is a regression method that can be used on data by considering spatial or location effects in the analyzed data. There are two types of spatial modeling, namely data modeling based on points and areas. Based on the area approach, spatial regression models can use several models, one of which is the Spatial Error Model (SEM). SEM is an error spatial model where there is a spatial correlation in the error. SEM considers that errors in regression models can be spatially correlated, which means that errors of observation in one location can affect errors in other locations. The SEM model was developed with two response variables called the Bivariate Spatial Error Model (BSEM). In this study, the BSEM method was applied to analyze the relationship between two variables, namely the Education Index and the Per Capita Expenditure Index. To evaluate the resulting model, this study uses the Akaike Information Criterion (AIC) and R-Square. The parameter assessment was carried out using Maximum Likelihood Estimation (MLE). Simultaneous testing was carried out using MLRT, the BSEM model parameter estimator was obtained using MLE with the Berndt-Hall-Hall-Hausman (BHHH) algorithm approach. The results of the study show that the BSEM model is good enough to model and predict the percentage of cases of the education index and the expenditure index in districts/cities in East Java in 2022. With the response variables that affect the model, namely the percentage of the poor population, population density, and the ratio of teachers to students and for the expenditure index are the percentage of the poor population, and population density.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	BSEM, MLE, MLRT, Indeks Pendidikan, Indeks Pengeluaran per-Kapita BSEM, MLE, MLRT, Education Index, per-Capita Expenditure Index.
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.7 Estimation H Social Sciences > HN Social history and conditions. Social problems. Social reform
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User:	Izzu Syifa'ian Alpain
Date Deposited:	05 Feb 2025 10:09
Last Modified:	05 Feb 2025 10:09
URI:	http://repository.its.ac.id/id/eprint/118263

Actions (login required)

View Item