Research Article

Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis

Volume: 14 Number: 2 December 31, 2024
EN TR

Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis

Abstract

The purpose of this research was to evaluate the predictive accuracy of various regression methods in the context of multiple linear regression when multicollinearity invalidates the underlying assumptions of the least squares method. These methods included least squares (LS), ridge regression (RR), lasso regression (LR), and principal component regression (PCR). For this aim, the dataset including 6 variables simulated from normal with different sample of size from range of 50 to 1000. The performance was assessed using mean square error (MSE) and R square value. Despite the existence of multicollienarity among independent variables, research findings showed that LS method had the smallest MSE in the training dataset but RR had the smallest mse in the test dataset. When the sample size increases, the mse values increase for each methods in the training set but decrease in the test set. They are closer to each other. In terms of R square values, all methods showed similar performance both training and test data set.

Keywords

References

  1. Alpar, R., (2017). Uygulamalı Çok Değişkenli İstatistiksel Yöntemler. Detay Yayıncılık.
  2. Altland H. W. (1999). Regression analysis: statistical modeling of a response variable.
  3. Çankaya, S., Eker, S., & Abacı, S. H. (2019). Comparison of least squares, ridge regression and principal component approaches in the presence of multicollinearity in regression analysis. Turkish Journal of Agriculture - Food Science and Technology, 7(3), 180-190. https://doi.org/10.24925/turjaf.v7i3.180-190.2019
  4. Chatterjee, S., & Hadi, A. S. (2006). Regression analysis by example. John Wiley & Sons.
  5. Draper, N. R. and Smith H. (1998). Applied Regression Analysis. New York, John Wiley and Sons, Inc.
  6. Fu, W. J. (1998). Penalized regression: the Bridge versus the Lasso, Journal of Computation and Graphical Statistics, 7, 397-416
  7. Göktaş, A., & Öznur, İ. (2010). Türkiye'de işsizlik oranının temel bileşenli regresyon analizi ile belirlenmesi. Sosyal Ekonomik Araştırmalar Dergisi, 10, 279-294.
  8. Gujarati, D. N., & Porter, D. C. (2009). Basic econometrics (5th ed.). McGraw-Hill.

Details

Primary Language

English

Subjects

Econometric and Statistical Methods

Journal Section

Research Article

Publication Date

December 31, 2024

Submission Date

September 13, 2024

Acceptance Date

December 29, 2024

Published in Issue

Year 2024 Volume: 14 Number: 2

APA
Ergişi, S., Doğanay, B., & Yavuz, Y. (2024). Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis. İstatistik Araştırma Dergisi, 14(2), 59-72. https://izlik.org/JA67EN45HU
AMA
1.Ergişi S, Doğanay B, Yavuz Y. Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis. JSRTR. 2024;14(2):59-72. https://izlik.org/JA67EN45HU
Chicago
Ergişi, Semih, Beyza Doğanay, and Yasemin Yavuz. 2024. “Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis”. İstatistik Araştırma Dergisi 14 (2): 59-72. https://izlik.org/JA67EN45HU.
EndNote
Ergişi S, Doğanay B, Yavuz Y (December 1, 2024) Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis. İstatistik Araştırma Dergisi 14 2 59–72.
IEEE
[1]S. Ergişi, B. Doğanay, and Y. Yavuz, “Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis”, JSRTR, vol. 14, no. 2, pp. 59–72, Dec. 2024, [Online]. Available: https://izlik.org/JA67EN45HU
ISNAD
Ergişi, Semih - Doğanay, Beyza - Yavuz, Yasemin. “Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis”. İstatistik Araştırma Dergisi 14/2 (December 1, 2024): 59-72. https://izlik.org/JA67EN45HU.
JAMA
1.Ergişi S, Doğanay B, Yavuz Y. Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis. JSRTR. 2024;14:59–72.
MLA
Ergişi, Semih, et al. “Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis”. İstatistik Araştırma Dergisi, vol. 14, no. 2, Dec. 2024, pp. 59-72, https://izlik.org/JA67EN45HU.
Vancouver
1.Semih Ergişi, Beyza Doğanay, Yasemin Yavuz. Performance Comparison of Least Squares, Ridge, Lasso and Principal Component Regression for Addressing Multicollinearity in Regression Analysis. JSRTR [Internet]. 2024 Dec. 1;14(2):59-72. Available from: https://izlik.org/JA67EN45HU