REGRESI KUANTIL UNTUK PEMODELAN TINGKAT PENGANGGURAN TERBUKA DI INDONESIA

NAVIANTI, DYNES RIZKY (2014) REGRESI KUANTIL UNTUK PEMODELAN TINGKAT PENGANGGURAN TERBUKA DI INDONESIA. Masters thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
1312201205 - Abstract.pdf - Published Version

Download (194kB) | Preview
[img] Text
1312201205 - Master Thesis.pdf - Published Version
Restricted to Repository staff only

Download (4MB) | Request a copy
[img]
Preview
Text
1312201205 - Conclusion.pdf - Published Version

Download (775kB) | Preview

Abstract

Prinsip metode kuadrat terkecil atau Ordinary Least Square (OLS) dalam analisis regresi yaitu meminimumkan jumlah kuadrat residual. Karena OLS fokus pada rata-rata sampel dan sensitif pada sebaran data yang asimetris dan terdapat pencilan, sehingga hasil estimasi parameter tidak stabil. Tujuan dari penelitian ini yaitu mendapatkan penduga parameter dari regresi kuantil dan menerapkan pada kasus riil pada data pengangguran di Indonesia. Problem pengangguran tersebut telah diselidiki bahwa terdapat kasus heteroskedastisitas apabila menerapkan metode regresi OLS. Penelitian dilakukan melalui kajian teoritis, studi simulasi, dan terapan. Hasil kajian teoritis menunjukkan bahwa penduga parameter regresi kuantil diperoleh dengan cara meminimumkan residual mutlak. Hasil studi simulasi menunjukkan bahwa regresi kuantil dapat mengatasi hadirnya pencilan dan asumsi homoskedastisitas yang tidak memenuhi pada regresi klasik. Selanjutnya, hasil kajian terapan pada kasus pengangguran di Indonesia menunjukkan bahwa hasil taksiran interval dari regresi kuantil lebih baik dan fleksibel. ===================================================================================================== The principle of the method of least squares or ordinary least squares (OLS) regression analysis is to minimize the sum of squared residuals. Because of OLS focus on the sample average and sensitive to the asymmetric distribution of the data and are outliers, it’s to be the unstable parameter estimates. The purpose of this study is to get the parameters of the quantile regression estimator and applying the real case on unemployment in Indonesia. The unemployment problem has been observed that there are cases when applying heteroscedasticity OLS regression method. The study was conducted through theoretical studies, simulation studies, and applied. The results of theoretical studies indicate that the quantile regression parameter estimators are obtained by minimizing the absolute residuals. The results of simulation studies show that quantile regression can cope with the presence of outliers and homoscedastic assumptions that do not meet the classical regression. Furthermore, the results of applied studies on unemployment in the Indonesian case shows that the estimated interval results from quantile regression more convenient and effective.

Item Type: Thesis (Masters)
Additional Information: RTSt 539.536 Nav r
Uncontrolled Keywords: Metode Kuadrat Terkecil, Regresi Kuantil, Tingkat Pengangguran Terbuka
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA278.2 Regression Analysis
Divisions: Faculty of Mathematics and Science > Statistics > (S2) Master Theses
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 05 Jan 2017 08:20
Last Modified: 05 Jan 2017 08:20
URI: http://repository.its.ac.id/id/eprint/1348

Actions (login required)

View Item View Item