Estimating Optimal Hodrick-Prescott Filter Smoothing Parameter for Turkey

This study aims at estimating the optimal smoothing parameter for Hodrick-Prescott filter for Turkey using 1987-2007 quarterly real Gross Domestic Product (GDP) data. Two alternative methods, proposed by Pedersen (2001) and Dermoune et al (2008) are used for estimation which gives the values for the HP Filter smoothing parameter for Turkey as 98 and 19, respectively. Using these estimates, we find that (1) estimated trend component of Turkish GDP is very volatile, which is in line with the conjectures of Aguiar and Gopinath (2007) for emerging market countries (2) absolute volatilities for the cyclical variations in GDP and its components are smaller compared to 1600 (3) our estimates give similar results with the frequently used smoothing parameter value in terms of relative volatility of components of GDP with respect to volatility of GDP.