Liu-Type Logistic Estimator


İnan D., Erdogan B.

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, cilt.42, ss.1578-1586, 2013 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 42
  • Basım Tarihi: 2013
  • Doi Numarası: 10.1080/03610918.2012.667480
  • Dergi Adı: COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.1578-1586
  • Anahtar Kelimeler: Liu-type estimator, Logistic regression, Multicollinearity, Ridge estimator
  • Marmara Üniversitesi Adresli: Evet

Özet

It is known that multicollinearity inflates the variance of the maximum likelihood estimator in logistic regression. Especially, if the primary interest is in the coefficients, the impact of collinearity can be very serious. To deal with collinearity, a ridge estimator was proposed by Schaefer et al. The primary interest of this article is to introduce a Liu-type estimator that had a smaller total mean squared error (MSE) than the Schaefer's ridge estimator under certain conditions. Simulation studies were conducted that evaluated the performance of this estimator. Furthermore, the proposed estimator was applied to a real-life dataset.