COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL

This study aimed to compare the robustness of the OLS method with a robust regression model on data that had outliers. The methods used on the robust regression model were M-estimation, MM-estimation, and S-estimation. The step taken was to check the characteristics of the data against outliers. Fur...

Full description

Saved in:
Bibliographic Details
Main Authors: Padrul Jana, Dedi Rosadi, Epha Diana Supandi
Format: Article
Language:English
Published: Universitas Pattimura 2023-06-01
Series:Barekeng
Subjects:
Online Access:https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/8057
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849762684451422208
author Padrul Jana
Dedi Rosadi
Epha Diana Supandi
author_facet Padrul Jana
Dedi Rosadi
Epha Diana Supandi
author_sort Padrul Jana
collection DOAJ
description This study aimed to compare the robustness of the OLS method with a robust regression model on data that had outliers. The methods used on the robust regression model were M-estimation, MM-estimation, and S-estimation. The step taken was to check the characteristics of the data against outliers. Furthermore, the data were modeled with and without outliers using the OLS method and the M-, MM-, and S-estimations. The results were very different between the data with and without the outlier models in the OLS method. It was reflected in the intercept and standard error variables generated from the models. Meanwhile, the regression model with the M-, MM-, and S-estimations was quite stable and able to withstand the presence of outliers. Based on the three estimations that were robust against the outliers, the MM-estimation was the best candidate because, in addition to having a stable intercept parameter estimation, it also had the smallest standard error, which was 61.9 in the resulting model.
format Article
id doaj-art-1460e16803e0451bbae20fcf7186a6cf
institution DOAJ
issn 1978-7227
2615-3017
language English
publishDate 2023-06-01
publisher Universitas Pattimura
record_format Article
series Barekeng
spelling doaj-art-1460e16803e0451bbae20fcf7186a6cf2025-08-20T03:05:41ZengUniversitas PattimuraBarekeng1978-72272615-30172023-06-011720979098810.30598/barekengvol17iss2pp0979-09888057COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODELPadrul Jana0Dedi Rosadi1Epha Diana Supandi2Department of Mathematics, FMIPA, Universitas Gadjah Mada, IndonesiaDepartment of Mathematics, FMIPA, Universitas Gadjah Mada, IndonesiaDepartment of Mathematics, FST, Universitas Islam Negeri Sunan Kalijaga Yogyakarta, IndonesiaThis study aimed to compare the robustness of the OLS method with a robust regression model on data that had outliers. The methods used on the robust regression model were M-estimation, MM-estimation, and S-estimation. The step taken was to check the characteristics of the data against outliers. Furthermore, the data were modeled with and without outliers using the OLS method and the M-, MM-, and S-estimations. The results were very different between the data with and without the outlier models in the OLS method. It was reflected in the intercept and standard error variables generated from the models. Meanwhile, the regression model with the M-, MM-, and S-estimations was quite stable and able to withstand the presence of outliers. Based on the three estimations that were robust against the outliers, the MM-estimation was the best candidate because, in addition to having a stable intercept parameter estimation, it also had the smallest standard error, which was 61.9 in the resulting model.https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/8057robust regressionm-estimationmm-estimations-estimation
spellingShingle Padrul Jana
Dedi Rosadi
Epha Diana Supandi
COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL
Barekeng
robust regression
m-estimation
mm-estimation
s-estimation
title COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL
title_full COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL
title_fullStr COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL
title_full_unstemmed COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL
title_short COMPARISON OF ROBUST ESTIMATION ON MULTIPLE REGRESSION MODEL
title_sort comparison of robust estimation on multiple regression model
topic robust regression
m-estimation
mm-estimation
s-estimation
url https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/8057
work_keys_str_mv AT padruljana comparisonofrobustestimationonmultipleregressionmodel
AT dedirosadi comparisonofrobustestimationonmultipleregressionmodel
AT ephadianasupandi comparisonofrobustestimationonmultipleregressionmodel