Identification of outlying and influential data with principal components regression estimation in binary logistic regression

Ozkale M. R.

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, cilt.50, sa.3, ss.609-630, 2021 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 50 Sayı: 3
  • Basım Tarihi: 2021
  • Doi Numarası: 10.1080/03610926.2019.1639749
  • Dergi Adı: COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Business Source Elite, Business Source Premier, CAB Abstracts, Compendex, Veterinary Science Database, zbMATH, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.609-630
  • Anahtar Kelimeler: Binary logistic regression, regression diagnostics, principal component logistic estimator, Monte Carlo simulation, Akaike's information criterion, DIAGNOSTICS
  • Çukurova Üniversitesi Adresli: Evet

Özet

In this study, we settle on the issue that when multicollinearity and unusual observations arise simultaneously and we straightforwardly extend leverages, Pearson residuals, delta beta and delta chi-square statistics using the principal components logistic regression (PCLR) estimator where the extensions typically take the advantage of the computation of PCLR estimator by one-step approximation. We then applied two simulation studies and a numerical example to illustrate the behavior of statistics for the PCLR estimator versus the traditional ML estimator.