Maharani, Annisa Dyah (2026) Pemodelan Tingkat Pengangguran Terbuka di Pulau Jawa Menggunakan Regresi Nonparametrik Spline Truncated. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5003221156-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5003221156-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Tingkat Pengangguran Terbuka (TPT) menggambarkan persentase angkatan kerja yang belum terserap dalam pasar kerja. Perbedaan kondisi ketenagakerjaan, pendidikan, kapasitas ekonomi, dan kebijakan pengupahan antarkabupaten/kota menyebabkan hubungan antara TPT dan faktor-faktor yang berkaitan dengannya tidak selalu membentuk pola tertentu. Penelitian ini bertujuan untuk membuat persebaran TPT beserta faktor-faktor yang diduga memengaruhinya serta menentukan faktor-faktor yang memengaruhi tinggi rendahnya TPT kabupaten/kota di Pulau Jawa tahun 2024 menggunakan regresi nonparametrik Spline Truncated. Data yang digunakan merupakan data sekunder Badan Pusat Statistik yang mencakup 118 kabupaten/kota, dengan TPT sebagai variabel respons serta Tingkat Partisipasi Angkatan Kerja (TPAK), Rata-Rata Lama Sekolah (RLS), Produk Domestik Regional Bruto (PDRB) per kapita, dan Upah Minimum Kabupaten/Kota (UMK) sebagai variabel prediktor. Hasil pemetaan menunjukkan bahwa TPT dan keempat variabel prediktor memiliki persebaran yang tidak merata antar kabupaten/kota. Rata-rata TPT sebesar 5,02%, dengan nilai minimum 1,56% dan maksimum 9,18%. Model terbaik diperoleh menggunakan kombinasi titik knot (3, 3, 3, 2), yaitu tiga titik knot pada TPAK, RLS, dan PDRB per kapita serta dua titik knot pada UMK. Model menghasilkan nilai Generalized Cross Validation minimum sebesar 1,227118 dan koefisien determinasi sebesar 71,55%. Pengujian secara serentak menunjukkan bahwa model signifikan, sedangkan pengujian secara parsial menunjukkan bahwa TPAK, RLS, PDRB per kapita, dan UMK masing-masing memiliki sedikitnya satu parameter yang signifikan. Model juga telah memenuhi asumsi residual identik, independen, dan berdistribusi normal. Hasil penelitian menunjukkan bahwa hubungan setiap variabel prediktor dengan TPT berubah pada interval yang dibatasi oleh titik knot, sehingga regresi nonparametrik Spline Truncated mampu menggambarkan pola hubungan TPT kabupaten/kota di Pulau Jawa tahun 2024.
=======================================================================================================================================
The Open Unemployment Rate (OUR) represents the percentage of the labor force that has not been absorbed into the labor market. Differences in labor market conditions, educational attainment, economic capacity, and wage policies across regencies and municipalities may cause the relationships between the OUR and its associated factors to follow varying patterns. This study aims to make the spatial distribution of the OUR and its associated factors and to determine the factors associated with the variation in the OUR across regencies and municipalities in Java Island in 2024 using truncated spline nonparametric regression. The study uses secondary data obtained from Statistics Indonesia, covering 118 regencies and municipalities. The OUR is used as the response variable, while the Labor Force Participation Rate (LFPR), Mean Years of Schooling (MYS), Gross Regional Domestic Product (GRDP) per capita, and Regency/Municipality Minimum Wage are used as predictor variables. The mapping results indicate that the OUR and the four predictor variables are unevenly distributed across regencies and municipalities. The average OUR was 5.02%, with a minimum of 1.56% and a maximum of 9.18%. The best model was obtained using the knot combination of (3, 3, 3, 2), consisting of three knot points for LFPR, MYS, and GRDP per capita and two knot points for the Regency/Municipality Minimum Wage. The model produced a minimum Generalized Cross-Validation value of 1.227118 and a coefficient of determination of 71.55%. The simultaneous significance test indicates that the model is statistically significant, while the partial significance tests show that LFPR, MYS, GRDP per capita, and the Regency/Municipality Minimum Wage each have at least one significant parameter. The model also satisfies the assumptions of homoscedasticity, independence, and normality of residuals. The results show that the relationship between each predictor variable and the OUR varies across intervals separated by the knot points. Therefore, truncated spline nonparametric regression is capable of representing the relationship patterns underlying the OUR across regencies and municipalities in Java Island in 2024.
Actions (login required)
 |
View Item |