Simulated Annealing Algorithm Based Ridge Estimator
Gazi University Journal of Science, cilt.39, sa.2, ss.780-792, 2026 (ESCI, Scopus, TRDizin)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 39 Sayı: 2
- Basım Tarihi: 2026
- Doi Numarası: 10.35378/gujs.1727816
- Dergi Adı: Gazi University Journal of Science
- Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus, TR DİZİN (ULAKBİM), Academic Search Ultimate (EBSCO), Biomedical Reference Collection: Corporate Edition (EBSCO), Engineering Source (EBSCO)
- Sayfa Sayıları: ss.780-792
- Anahtar Kelimeler: Biasing parameter, Multicollinearity, Optimization, Ridge regression, Simulated annealing
- Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
- Gazi Üniversitesi Adresli: Evet
Özet
Multicollinearity is a significant problem in multiple linear regression. Different researchers have suggested biased estimators as a possible solution to address the issue of multicollinearity, and an example of a biased estimator is the ridge regression estimator. Estimating the bias parameter is an essential problem for the ridge regression estimator. This paper presents a new solution method that utilizes simulated annealing optimization to determine the optimal bias parameter as an alternative to the ridge regression bias value proposed by Hoerl and Kennard. We obtained the bias parameter estimation values using the proposed solution method, considering various dependency structures, sample sizes, variance, and number of variables.