Print ISSN: 1680-855X

Online ISSN: 2664-2956

Keywords : Leverage Points


Robust Weighted Approaches to Detect and Deal with Outliers in Estimating Principal Component Regression Model

Esraa Najeeb Alsaraf; Bashar Abdulaziz AL-Talib

IRAQI JOURNAL OF STATISTICAL SCIENCES, 2021, Volume 18, Issue 33, Pages 1-21
DOI: 10.33899/iqjoss.2021.168371


Abstract
This paper aims to propose an approach to deal with the problem of Multi-Collinearity between the explanatory variables and outliers in the data by using the method of Principal Component Regression, and then using a robust weighting functions for the objective function has been used to deal with the presence of outliers in the data, and in order to verify the efficiency of the estimators, an experimental study was conducted through the simulation approach, and the methods were also applied to real data collected from the files of Badoush Cement Factory in Nineveh Governorate for the period from (2008-2014) with nine explanatory variables representing the chemical properties of cement and a dependent variable representing the physical properties of cement (hardness).
The data was tested whether it was suffer from multi-collinearity problem and then the least squares using principal components as an explanatory variables and the model was estimated, and it was found that the variables suffer from Multi-Collinearity problem, and the treatment was done by applying principal component regression weighed by robust weights due to the presence of outlying values in the data in addition to the collinearity problem.

The Effect of the Outliers and Leverage Points in the Construction of the Bayesian and Bootstrap Confidence Intervals

Muzahim Mohammed

IRAQI JOURNAL OF STATISTICAL SCIENCES, 2019, Volume 16, Issue 28, Pages 77-110
DOI: 10.33899/iqjoss.2019.164184

The aim of this research is to compare the bootstrap confidence intervals with the Bayesian confidence intervals for smoothing splines as well as the traditional confidence intervals to determine which of these limits are best in the presence of Outliers and Leverage points in data. The simulation experiments were conducted on two models: the first was linear in the presence of data that was contaminated with outliers and the other with the Leverage points: The second model was nonlinear in the presence of data contaminated with outlying observations 
Simulation experiments were carried out on different samples. The Penalized Least Squares method was used to fit the Nonparametric regression. The Generalized Cross Validation function (GCV) was used to select the amount of smoothing.