A comparative study of estimation methods for parameters in multiple linear regression model


CANKAYA S., KAYAALP G. , Sangun L. , TAHTALI Y., AKAR M.

JOURNAL OF APPLIED ANIMAL RESEARCH, cilt.29, ss.43-47, 2006 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 29 Konu: 1
  • Basım Tarihi: 2006
  • Doi Numarası: 10.1080/09712119.2006.9706568
  • Dergi Adı: JOURNAL OF APPLIED ANIMAL RESEARCH
  • Sayfa Sayıları: ss.43-47

Özet

This paper investigated least squares method, non-parametric method and robust regression methods to predict the parameters of multiple regression models. To evaluate these methods, measurements of body weight, total length and fork length of fishes collected from Serranus cabrilla were used. In these regression models, body weight was dependent variable whereas total length and fork length were independent variables. The results show that non-parametric regression method, general additive model, has minimum R-2 value and least median squares has maximum R-2 value, 0.334 and 0.855, respectively.