Modified ridge‐type estimator to combat multicollinearity: Application to chemical data. (7th April 2019)

• Record Type: Journal Article
• Authors: Lukman, Adewale F.; Ayinde, Kayode; Binuomote, Samuel; Clement, Onate A.
• Is Part Of: Journal of chemometrics. Volume 33:Number 5(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-04-07
• Subjects: modified ridge‐type estimator -- Monte Carlo simulation -- multicollinearity -- ridge regression estimator; Chemistry -- Mathematics -- Periodicals; Chemistry -- Statistical methods -- Periodicals; 542.85
• DOI: 10.1002/cem.3125
• Languages: English
• ISSNs: 0886-9383

Abstract

Abstract: The Linear regression model is one of the most widely used models in different fields of study. The most popularly used estimation technique is the ordinary least squares estimator. The technique becomes unstable and gives misleading result in the presence of multicollinearity. The ridge regression estimator has been widely accepted as an alternative method to combat the problem of multicollinearity. In this study, a modified ridge‐type estimator is suggested by modifying the ridge regression estimator. A Monte Carlo simulation study and real‐life application were conducted to compare the performance of this estimator and some other existing estimators. The results of both simulation study and real‐life application show that the proposed estimator outperforms other competing estimator. Abstract : This study provides an alternative method of estimation in the presence of multicollinearity.