Pemodelan PDRB Provinsi di Jawa pada Jumlah UMK dan Faktor Lain dengan Regresi Robust Least Trimmed Square dan Estimasi S

Zahra, Alivia (2021) Pemodelan PDRB Provinsi di Jawa pada Jumlah UMK dan Faktor Lain dengan Regresi Robust Least Trimmed Square dan Estimasi S. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 06211740000048-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (1MB) | Request a copy

Abstract

Salah satu indikator untuk mengetahui pertumbuhan ekonomi suatu daerah adalah PDRB. Faktor-faktor yang diduga berpengaruh terhadap PDRB yaitu Jumlah UMK, IPM, Rasio Ketergantungan, Pengeluaran per Kapita, Rata-ata Lama Sekolah, dan Upah Minimum Kab/Kota. Hasil Sensus Ekonomi tahun 2016 menunjukkan, Jawa masih menjadi konsentrasi UMK. Kondisi ini harus dioptimalkan karena akan berdampak kepada pertumbuhan ekonomi di Indonesia. Pada penelitian ini hasil deteksi menggunakan DFFIT menunjukkan terdapat data outliers. Apabila terdapat pengamatan outlier maka metode OLS tidak dapat bekerja dengan baik dan mengganggu pemenuhan asumsi dan estimasi parameter, sehingga penggunaan regresi robust dilakukan untuk menangani hal tersebut. Metode regresi robust Estimasi S dan LTS (Least Trimmed Square) merupakan metode regresi robust yang baik karena memiliki breakdown point yang tinggi. Hasil analisis kajian simulasi menunjukkan metode regresi robust lebih baik dari regresi OLS, regresi LTS secara umum lebih baik dari regresi robust estimasi S karena menghasilkan MSE dan standar error terendah. Demikian pula hasil analisis terapan studi kasus PDRB yang memberikan hasil tak jauh berbeda. Hasil analisis studi terapan menunjukkan, Jumlah UMK (x1) signifikan berpengaruh positif terhadap PDRB (y) pada model Provinsi Jawa Timur, Jawa Tengah, dan DI. Yogyakarta. Indeks Pembangunan Manusia (x2) signifikan berpengaruh positif terhadap PDRB(y) pada model Jawa Tengah, DI. Yogyakarta, Jawa Barat, Banten, dan DKI. Jakarta. Upah Minimum Kab/ Kota (x6) signifikan berpengaruh positif terhadap PDRB(y) di Jawa Timur.
=================================================================================================
One indicator to determine the economic growth of a region is to look at the GDRP. Factors that are thought to influence GDRP are the number of Micro and Small Enterprises (MSEs), HDI, Dependency Ratio, Expenditure per Capita, Average Years of Schooling, and District/City Minimum Wage. The results of the 2016 Economic Census show that Java is still a concentration of MSEs. This condition must be optimized because it will have an impact on economic growth in Indonesia. In this study, the results of detection using DFFIT showed that there were outliers. If there are outliers in the data analysis, the OLS method cannot work properly and interfere with the fulfillment of assumptions and parameter estimates, so that the use of robust regression is carried out to handle this. Robust S Estimation and LTS robust regression methods are good robust regression methods because they have high breakdown points. The results of the analysis of the simulation study show that the robust regression method is better than the OLS regression, LTS regression is generally better than S Estimation robust regression because it produces the lowest MSE and standard error. Otherwise, the results of the applied analysis of the GDRP case study which gave results were not much different than the simulation. The results shows that the number of MSEs (x1) has a significant positive effect on GDRP (y) in the East Java, Central Java, and DI. Yogyakarta models. The Human Development Index (x2) has a significant positive effect on GDRP(y) in the Central Java, DI. Yogyakarta, West Java, Banten, and DKI. Jakarta. Distric/City Minimum Wage (x6) has a significant positive effect on GDRP(y) in East Java model.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Estimasi S, Least Trimmed Square, Ordinary Least Square, PDRB, Regresi Robust, UMK, GDRP, MSEs, Robust Regression, S Estimation
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics H Social Sciences > HA Statistics > HA31.3 Regression. Correlation H Social Sciences > HD Industries. Land use. Labor > HD1393.25 Business enterprises Q Science
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Alivia Zahra
Date Deposited:	01 Sep 2021 01:22
Last Modified:	01 Sep 2021 01:22
URI:	http://repository.its.ac.id/id/eprint/91435

Actions (login required)

View Item