Estimation of GDP in Turkey by nonparametric regression models

AYDIN D.

6th WSEAS International Conference on Instrumentation, Measurement, Circuits and Systems/7th WSEAS International Conference on Robotics, Control and Manufacturing Technology, Hangzhou, China, 15 - 17 April 2007, pp.221-222 identifier

  • Publication Type: Conference Paper / Full Text
  • City: Hangzhou
  • Country: China
  • Page Numbers: pp.221-222
  • Keywords: estimation, semi-parametric models, additive models, smoothing spline, trend
  • Anadolu University Affiliated: Yes

Abstract

Present study is about using of nonparametric models for GDP (Gross Domestic Product) per capita prediction in Turkey. It has been considered two alternative situations due to seasonal effects. In the first case, it is discussed a semi-parametric model where parametric component is dummy variable for the seasonality. In the second case, it is considered the seasonal component to be a smooth function of time, and therefore, the model falls within the class of additive models. The results obtained by semi-parametric regression models are compared to those obtained by additive nonparametric and parametric linear models.