PEMODELAN FAKTOR-FAKTOR YANG MEMPENGARUHI TINGKAT KEMISKINAN DI INDONESIA MENGGUNAKAN ELASTIC NET REGRESSION DENGAN OPTIMASI HYPERPARAMETER OPTUNA

Nur, Yunita (2025) PEMODELAN FAKTOR-FAKTOR YANG MEMPENGARUHI TINGKAT KEMISKINAN DI INDONESIA MENGGUNAKAN ELASTIC NET REGRESSION DENGAN OPTIMASI HYPERPARAMETER OPTUNA. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21083010107_Cover.pdf Download (2MB)
	Text (Bab 1) 21083010107_Bab 1.pdf Download (244kB)
	Text (Bab 2) 21083010107_Bab 2.pdf Restricted to Repository staff only until 25 July 2028. Download (746kB)
	Text (Bab 3) 21083010107_Bab 3.pdf Restricted to Repository staff only until 25 July 2028. Download (557kB)
	Text (Bab 4) 21083010107_Bab 4.pdf Restricted to Repository staff only until 25 July 2028. Download (2MB)
	Text (Bab 5) 21083010107_Bab 5.pdf Restricted to Repository staff only until 25 July 2028. Download (181kB)
	Text (Daftar Pustaka) 21083010107_Daftar Pustaka.pdf Download (163kB)
	Text (Lampiran) 21083010107_Lampiran.pdf Restricted to Repository staff only until 25 July 2028. Download (181kB)

Abstract

Poverty in Indonesia is a complex problem and is influenced by various interconnected aspects of community welfare. This study aims to evaluate the factors that influence poverty levels using the Elastic Net Regression method optimized through Optuna. This method was chosen based on its ability to overcome multicollinearity between variables and produce a more consistent prediction model. The data used in this study comes from secondary sources owned by the Central Statistics Agency (BPS) and includes eight independent variables, namely Life Expectancy (AHH), Literacy Rate (AMH), Gross Regional Domestic Product (GRDP), Human Development Index (HDI), Labor Force Participation (PAK), Average Years of Schooling (RLS), Construction Cost Index (IKK), and Early Marriage. The modeling results show that the Elastic Net optimized with Optuna provides the best performance, with an R² value of 0.7155, MAE of 3.1404, and MSE of 16.1331, and has met all classical regression assumptions. The results revealed six variables significantly influencing poverty levels: the Human Development Index (HDI), the Age of Childhood, the Age of Childhood, and Early Marriage, all of which had negative effects, and the Gender-Based Social Welfare Index (RLS) and the Age-Based Social Welfare Index (PAK), which had positive effects. The study's conclusions underscore the importance of strengthening human development, improving education quality, and controlling early marriage as part of a more effective and sustainable poverty reduction strategy in Indonesia.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Hindrayani, Kartika Maulida	NIDN0009099205	kartikamaulida.ds@upnjatim.ac.id
Thesis advisor	Damaliana, Aviolla Terza	NIDN0002089402	aviolla.terza.sada@upnjatim.ac.id

Subjects:

H Social Sciences > HA Statistics
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Yunita Nur

Date Deposited:

28 Jul 2025 07:33

Last Modified:

28 Jul 2025 07:33

URI:

https://repository.upnjatim.ac.id/id/eprint/40866