It is known that parametric and nonparametric methods are used for nonlinear time series. Of the parametric methods, autoregressive (AR) model and self-threshold value (SETAR) model and, of the nonparametric methods, additive regression model (ARM) have been used in this study. Nonparametric regression techniques are often sensitive to presence of otocorrelation in errors. Practical results of this sensitivity is explanied by appropriate selection of smoothing parameter. In this context, backfittting algorithm based on smoothing spline method in the existing literature is discussed. As an application, an appropriate model for the export unit value index data for Turkey is try to be determined by fitting each of AR, SETAR and, ARM models to the data.
Alan : Sosyal, Beşeri ve İdari Bilimler
Dergi Türü : Ulusal
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|