Modified Ridge Logistic Estimator Based on Singular Value Decomposition | ||||
The Egyptian Statistical Journal | ||||
Article 2, Volume 67, Issue 1, June 2023, Page 19-35 PDF (660.05 K) | ||||
Document Type: Original Article | ||||
DOI: 10.21608/esju.2023.206494.1012 | ||||
View on SCiNiTO | ||||
Authors | ||||
Mostafa Kamal 1; Monira Ahmed Hussein2 | ||||
1مدرس مساعد بكلية التجارة - جامعة مدينة السادات | ||||
2أستاذ الإحصاء المتفرغ- بكلية التجارة - جامعة مدينة السادات | ||||
Abstract | ||||
This paper aims to introduce a modification of the ridge estimator based on the singular value decomposition (SVD) technique of the design matrix (X ) to combat multicollinearity in the binary logistic model. This estimator is called a modified ridge logistic based on SVD estimator which is denoted as (MRLSVDE). We study the statistical properties of the proposed estimator in the context of the bias, variance-covariance matrix, and mean squared error. A Monte Carlo simulation study is conducted to evaluate the performance of the proposed estimator over the ridge logistic estimator (RLE) and the maximum likelihood estimator (MLE) based on the scalar mean squared error (SMSE) criterion. Moreover, an empirical application is provided to investigate the potential benefits of the proposed estimator in real-life fields. The results of the simulation study and real data application indicate that the proposed estimator outperforms the maximum likelihood and ridge logistic estimators in the scalar mean squared error sense. | ||||
Keywords | ||||
Logistic regression model; Multicollinearity; Ridge logistic estimator; Singular value decomposition; Numerical rank | ||||
Statistics Article View: 83 PDF Download: 205 |
||||