Comparison between Linear Regression and Robust Regression Models by Using Error Criteria and Information Criteria Applied to Human Sample

Hypertension is a common and serious disease, and for this reason, a sample of patients was chosen from Azadi Teaching Hospital in Duhok. In this study a comparison was made between Ordinary Least Squares (OLS) with two robust methods Least Trimmed Square Estimator (LTS) and the Modified Maximum li...

Full description

Saved in:
Bibliographic Details
Main Author: Ismat Mousa Ibrahim
Format: Article
Language:Arabic
Published: Salahaddin University-Erbil 2025-04-01
Series:Zanco Journal of Humanity Sciences
Subjects:
Online Access:https://zancojournal.su.edu.krd/index.php/JAHS/article/view/2446
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Hypertension is a common and serious disease, and for this reason, a sample of patients was chosen from Azadi Teaching Hospital in Duhok. In this study a comparison was made between Ordinary Least Squares (OLS) with two robust methods Least Trimmed Square Estimator (LTS) and the Modified Maximum likelihood type estimator (MM), they were evaluated using two types of criteria represented by error criteria (MSE, MAPE, MSAE, SAZ1, SAZ2,) and information criteria (AIC, BIC). Criteria have played a fundamental role in the statistics field and at the same time have applied to obtain the lowest rate of errors as well as to get the optimal solutions. The Modified Maximum likelihood type estimator method showed high efficiency in calculating values by most criteria, with exception of the SAS2 criteria, which recorded as the best value using the Least Trimmed Square Estimator method, and the (MSE) criteria, which showed best value using the Ordinary least squares method.”
ISSN:2412-396X