Detection of influential case in fuzzy regression

Basaran A., Gunay S.

2nd International Conference on Soft Methods in Probability and Statistics (SMPS 2004), Oviedo, İspanya, 02 Eylül 2004 - 04 Eylül 2007, ss.439-446 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Basıldığı Şehir: Oviedo
  • Basıldığı Ülke: İspanya
  • Sayfa Sayıları: ss.439-446
  • Hacettepe Üniversitesi Adresli: Hayır

Özet

Influential case has not been addressed yet in the fuzzy linear regression. The Influential case is an outlier which causes to change the estimation of the parameters of the model, when it is omitted. We proposed an approach here to detect the influential case in the fuzzy linear regression. A simple example is given to illustrate a problem using the fuzzy least squares regression and then we use two approaches to compare the fuzzy numbers obtained from the predicted model based on the available data with the fuzzy numbers predicted by the models which are obtained by omitting each case.