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.

Preview	Text 1312201205 - Abstract.pdf - Published Version Download (194kB) | Preview
	Text 1312201205 - Master Thesis.pdf - Published Version Restricted to Repository staff only Download (4MB) | Request a copy
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. Logistic regression
Divisions:	Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis
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