skip to main content

REGRESI ROBUST ESTIMASI-M DENGAN PEMBOBOT ANDREW, PEMBOBOT RAMSAY DAN PEMBOBOT WELSCH MENGGUNAKAN SOFTWARE R

*Aulia Desy Deria  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Abdul Hoyyi  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Mustafid Mustafid  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Open Access Copyright 2020 Jurnal Gaussian under http://creativecommons.org/licenses/by-nc-sa/4.0.

Citation Format:
Abstract

Robust regression is one of the regression methods that robust from effect of outliers. For the regression with the parameter estimation used Ordinary Least Squares (OLS), outliers can caused assumption violation, so the estimator obtained became bias and inefficient. As a solution, robust regression M-estimation with Andrew, Ramsay and Welsch weight function can be used to overcome the presence of outliers. The aim of this study was to develop a model for case study of poverty in Central Java 2017 influenced by the number of unemployment, population, school participation rate, Human Development Index (HDI), and inflation. The result of estimation using OLS show that there is violation of heteroskedasticity caused by the presence outliers. Applied robust regression to case study proves robust regression can solve outliers and improve parameter estimation. The best robust regression model is robust regression M-estimation with Andrew weight function. The influence value of predictor variables to poverty is 92,7714% and MSE value is 370,8817.

 

Keywords: Outliers, Robust Regression, M-Estimator, Andrew, Ramsay, Welsch

Fulltext View|Download
Keywords: Outliers, Robust Regression, M-Estimator, Andrew, Ramsay, Welsch

Article Metrics:

Last update:

No citation recorded.

Last update:

No citation recorded.