REGRESI KOMPONEN UTAMA ROBUST S-ESTIMATOR UNTUK ANALISIS PENGARUH JUMLAH PENGANGGURAN DI JAWA TENGAH

*Jeffri Nelwin J. O. Siburian  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Rita Rahmawati  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Abdul Hoyyi  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Published: 29 Nov 2019.
Open Access Copyright 2020 Jurnal Gaussian
License URL: http://creativecommons.org/licenses/by-nc-sa/4.0

Citation Format:
Abstract

Robust principal component regression s-estimator is principal component regression that applies robust approach method at principal component analysis and s-estimator at principal component regression analysis. The aim of robust principal component regression s-estimator is to overcome multicollinearity problems in multiple linier regression Ordinary Least Square (OLS) and to overcome outlier problems in principal component regression so get the most effective model. Minimum Volume Ellipsoid (MVE) is one of the robust approach methods that applied when doing principal component analysis and S-Estimator is one of the estimation methods that applied when doing principal component regression analysis. The case in this study is the factors that influence the Number of Unemployment in Central Java in 2017. The model that provides the most effective result to handling multicolliniearity and ouliers in the case study  Number of Unemployment in Central Java in 2017 is using robust principal component regression MVE-(S-Estimator) with Adjusted R2 value of 0.9615 and RSE value of 0.4073.

 

Keywords: Robust Principal Component Regression S-Estimator, Multicollinearity, Outliers, Minimum Volume Ellipsoid (MVE), Number of Unemployment.
Keywords: Robust Principal Component Regression S-Estimator, Multicollinearity, Outliers, Minimum Volume Ellipsoid (MVE), Number of Unemployment

Article Metrics: