Azizah, Salma (2024) Pemodelan Indeks Keparahan Kemiskinan di Indonesia Menggunakan Regresi Probit Biner dengan Efek Interaksi. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
5003201180-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (3MB) | Request a copy |
Abstract
Indeks Keparahan Kemiskinan merupakan pengukuran yang menggambarkan penyebaran pengeluaran diantara penduduk miskin yang berada di bawah garis kemiskinan. Semakin tinggi nilai indeks, maka ketimpangan pengeluaran yang terjadi diantara penduduk miskin semakin tinggi pula. Terjadinya ketimpangan tersebut mengindikasikan adanya ketidakmerataan dalam upaya peningkatan kesejahteraan antar penduduk miskin. Angka Indeks Keparahan Kemiskinan di Indonesia sendiri dari tahun ke tahun menunjukkan perkembangan yang fluktuatif. Tetapi jika dilihat capaian Indeks Keparahan Kemiskinan secara nasional dari tahun 2021 ke tahun 2022 menunjukkan penurunan angka. Demi mencegah adanya kenaikan angka tersebut, maka perlu dilakukan analisis lebih lanjut mengenai faktor-faktor yang diduga mempengaruhi Indeks Keparahan Kemiskinan. Dalam penelitian ini, metode yang digunakan yaitu regresi probit biner dengan efek interaksi. Variabel respon Indeks Keparahan Kemiskinan akan dibagi menjadi dua kategori yaitu Indeks Keparahan Kemiskinan dibawah nasional dan Indeks Keparahan Kemiskinan diatas nasional. Dari hasil uji interaksi diperoleh satu pasang variabel prediktor yang saling berinteraksi yaitu Tingkat Partisipasi Angkatan Kerja dengan persentase penduduk 15 tahun keatas yang tamat perguruan tinggi. Sehingga, dalam pemodelan akan melibatkan enam variabel prediktor efek utama dan satu variabel interaksi. Setelah dilakukan pemodelan menggunakan regresi probit biner dengan efek interaksi, variabel prediktor yang berpengaruh signifikan terhadap Indeks Keparahan Kemiskinan diantaranya yaitu Tingkat Partisipasi Angkatan Kerja, persentase penduduk 15 tahun keatas yang tamat perguruan tinggi, persentase Rumah Tangga penerima program Bantuan Pangan Non Tunai, serta variabel interaksi antara Tingkat Partisipasi Angkatan Kerja dengan persentase penduduk 15 tahun keatas yang tamat Perguruan Tinggi. Ketepatan klasifikasi dari hasil prediksi model mencapai 79,41%.
=====================================================================================================================================
The poverty severity index is a measurement that describes the distribution of expenditure among poor people who are below the poverty line. The higher index value showed the higher the expenditure inequality that occurs among the poor. The occurrence of this inequality indicates that there is inequality in efforts to improve welfare among the poor. The poverty severity index figures in Indonesia show fluctuating developments from year to year. However, if we look at the achievements of the national poverty severity index from 2021 to 2022, the figures show a decline. In order to prevent an increase in this number, it is necessary to carry out further analysis regarding the factors that are thought to influence the poverty severity index. In this research, the method used is binary probit regression with interaction effects. The poverty severity index response variable will be divided into two categories, namely the poverty severity index below the national level and the poverty severity index above the national level. From the results of interaction test, one pair of predictor variables that interact with each other is obtained, namely the labor force participation level and the percentage of the population 15 years and over who have graduated from college. So, the modelling will involve six main effect predictor variables and one interaction variable. After modelling using binary probit regression with interaction effects, the predictor variables that have a significant effect on the poverty severity index include the labor force participation rate, the percentage of the population 15 years and over who have graduated from college, the percentage of households receiving the Non-Cash Food Assistance program, as well as interaction variables between the labor force participation rate and the percentage of the population 15 years and over who have graduated from college. The classification accuracy of the model prediction results reached 79,41%.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | binary probit regression, interaction effects, poverty severity index, regresi probit biner, efek interaksi, indeks keparahan kemiskinan |
Subjects: | H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
Depositing User: | Salma Azizah |
Date Deposited: | 12 Aug 2024 05:24 |
Last Modified: | 29 Aug 2024 02:27 |
URI: | http://repository.its.ac.id/id/eprint/114952 |
Actions (login required)
View Item |