Comparision of Principal Component Regression with the Least Square Method in Prediction of Internal Egg Quality Characteristics in Japanese Quails

Rathert T. C., Uckardes F., NARİNÇ D., AKSOY T.

KAFKAS UNIVERSITESI VETERINER FAKULTESI DERGISI, cilt.17, sa.5, ss.687-692, 2011 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 17 Sayı: 5
  • Basım Tarihi: 2011
  • Dergi Adı: KAFKAS UNIVERSITESI VETERINER FAKULTESI DERGISI
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.687-692
  • Anahtar Kelimeler: Principal component regression, Multicollinearity, Least square method, Japanese quail, COTURNIX-COTURNIX-JAPONICA, GENETIC-PARAMETERS, RIDGE-REGRESSION, DIFFERENT LINES, CARCASS TRAITS, BODY-WEIGHT
  • Akdeniz Üniversitesi Adresli: Evet

Özet

The purpose of this study is to determine the inner egg quality characteristics albumen height, albumen width, albumen length, yolk diameter and yolk height using principal component regression. For this reason 104 eggs of Japanese quails between the 20(th) and 24(th) week of age were analysed. The birds had not been object of selection and they were paired randomly. The problem of multicollinearity occurs in regression models between independent variables in case of high correlation. In case of multicollinearity, parameter estimations and hypothesis results contradict each other through large standard deviation caused by Least Square method (LS). One of the methods that are applied in such a case is principal component regression (PCR). PCR leads to small standard deviation and more accurate and more reliable regression equations. For this research the external egg quality traits egg weight (X(1)), egg width (X(2)), egg length (X(3)) and shape index (X(4)) were used as variables. By processing these variables using LS and PCR the inner egg quality traits albumen height, albumen width, albumen length, yolk diameter and yolk height were estimated. In either method the regression estimating equations of the inner egg quality were significant (P < 0.01). The goodness of fit of the regression estimating equations was 29.24%-68.25% when LS and 29.12%-66.72% when PCR was applied.