ANALISIS HASIL MODEL REGRESI GENERALIZED POISSON, QUASI-POISSON, DAN BINOMIAL NEGATIF PADA ANGKA KEMISKINAN DI PROVINSI JAWA BARAT

OKTAVIA ANGGI PRATIWI, . (2023) ANALISIS HASIL MODEL REGRESI GENERALIZED POISSON, QUASI-POISSON, DAN BINOMIAL NEGATIF PADA ANGKA KEMISKINAN DI PROVINSI JAWA BARAT. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (1MB)
	Text BAB 1.pdf Download (241kB)
	Text BAB 2.pdf Restricted to Registered users only Download (415kB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (312kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (473kB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (248kB) | Request a copy
	Text DAFTAR PUSTAKA.pdf Download (226kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (825kB) | Request a copy

Abstract

Model regresi yang paling sederhana untuk menganalisis data yang melibatkan variabel respon dalam bentuk data cacahan adalah regresi Poisson. Namun, sering kali data rill di lapangan mengalami kondisi overdispersi. Untuk mengatasi situasi ini, alternatif pendekatan yang bisa digunakan adalah regresi Generalized Poisson, quasi-Poisson, dan binomial negatif. Pada penelitian ini, akan diterapkan model regresi optimal/terbaik yang sesuai dalam memodelkan jumlah penduduk miskin di Provinsi Jawa Barat yang merupakan data cacahan dengan kondisi overdispersi. Perbandingan model antara regresi Generalized Poisson, quasi-Poisson, dan binomial negatif dilakukan dengan melihat hasil rasio deviance dan hasil RMSE. Hasil penelitian menunjukkan bahwa rasio antara deviance dengan derajat bebas berdasarkan regresi Generalized Poisson, regresi Quasi-Poisson, dan regresi Binomial Negatif masing masing adalah 0,732 ; 4514,797 ; dan 1,700, sedangkan hasil perhitungan RMSE dari ketiga model tersebut adalah 0,745 ; 50,082 ; dan 0,971. Hasil perbandingan menunjukkan bahwa regresi Generalized Poisson merupakan model yang paling sesuai dalam memodelkan jumlah penduduk miskin di Provinsi Jawa Barat. Berdasarkan hasil pengujian hipotesis pada taraf nyata 5% membuktikan bahwa kepadatan penduduk (X1), laju perekonomian atas dasar harga konsumen (X2), harapan lama sekolah (X6), jumlah sekolah (X12), persentase penerima jaminan kesehatan (X14), dan jumlah kejadian bencana alam (X16) merupakan faktor-faktor yang berpengaruh nyata dan signifikan terhadap jumlah penduduk miskin di Provinsi Jawa Barat. ************* The simplest regression model for analyzing data involving response variables in the form of countable data is Poisson regression. However, often real data in the field experience overdispersion conditions. To overcome this situation, alternative approaches that can be used are Generalized Poisson, quasi-Poisson, and negative binomial regression. In this study, the optimal/best regression model will be applied which is appropriate in modeling the number of poor people in West Java Province which is an overdispersion of enumerated data. Model comparison between Generalized Poisson, quasi-Poisson, and negative binomial regressors was carried out by looking at the results of the deviance ratio and the resultsl RMSE. lresearch results show that the ratio between deviance and degrees of freedom based on regression Generalized Poisson, regression Quasi-Poisson, and the Negative Binomial regression are 0.732 ; 4514,797 ; and 1.700, while the result of calculating the RMSE of the three models is 0.745 ; 50,082 ; and 0.971. The results of the comparison show that Generalized Poisson regression is the most appropriate model in modeling the number of poor people in West Java Province. Based on the testing resultslthe hypothesis at the 5% level of significance proves that population density (X1), economic growth on the basis of consumer prices (X2), expected length of schooling (X6), number of schools (X12), percentage of health insurance recipients (X14), and number of natural disasters (X16) are factors that have a real and significant effect on the number of poor people in West Java Province.

Item Type:	Thesis (Sarjana)
Additional Information:	1) Vera Maya Santi, M.Si. 2) Ibnu Hadi, M.Si.
Subjects:	Sains > Matematika
Divisions:	FMIPA > S1 Matematika
Depositing User:	Users 19212 not found.
Date Deposited:	06 Sep 2023 05:44
Last Modified:	06 Sep 2023 05:44
URI:	http://repository.unj.ac.id/id/eprint/40941

Actions (login required)

View Item