Econometrics application of partial least squares regression: an endogeneous growth model for Turkey

Korkmazoglu O. B., KEMALBAY G.

World Conference on Business, Economics and Management (BEM), Antalya, Türkiye, 4 - 06 Mayıs 2012, cilt.62, ss.906-910 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 62
  • Doi Numarası: 10.1016/j.sbspro.2012.09.153
  • Basıldığı Şehir: Antalya
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.906-910
  • Anahtar Kelimeler: Partial least squares, multicollinearity, economic growth, PLS
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

Many Econometric models which included time series data have multicollinearity problem. Partial least square regression (PLS) is one of the popular multivariate regression methods in a a wide range of fields.The reason of that PLS have been designed to confront the situation that many correlated predictor variables and few samples situation. Growth rate is determined endogenously in endogenous growth models. In this study different algoritms (like Kernel, NIPALS, etc) of PLS are applied to an endogenous growth model starting with works of Romer (1986) and Lucas (1988). We study on real data for Turkey to illustrate the econometric applications and interpretations of various PLS algoritms using R programming. (C) 2012 Published by Elsevier Ltd. Selection and/or peer review under responsibility of Prof. Dr. Huseyin Arasli