VISUALIZING DIAGNOSTIC OF MULTICOLLINEARITY: TABLEPLOT AND BIPLOT METHODS

Alkan B. B., ATAKAN C.

PAKISTAN JOURNAL OF STATISTICS, cilt.29, sa.1, ss.59-78, 2013 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 29 Sayı: 1
  • Basım Tarihi: 2013
  • Dergi Adı: PAKISTAN JOURNAL OF STATISTICS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.59-78
  • Anahtar Kelimeler: Multicollinearty, Biplot, Tableplot, Visualizing approach, RIDGE REGRESSION, MATRICES
  • Akdeniz Üniversitesi Adresli: Hayır

Özet

In this study we tried to bring out the importance of two methods used for visualizing multicollinearity diagnostics. The first method is called Tableplot (see Friendly and Kwan, 2009). Through this technique, multicollinearity diagnostics ideas are combined with a visualizing approach. The second method is called Biplot (see Gabriel, 1971; Gower and Hand, 1996). The biplot technique is used for demonstrating significant properties of multivariate data structure. Both of these approaches have been examined by one real and three artificial data. Results indicate that Biplot method is preferred to instead of Tableplot for too large condition indexes.