Estimating Potential Output in Romania using Univariate Statistical Filters

In this paper, we determine the potential economic growth in Romania using three statistical univariate filters. The techniques used are Hodrick-Prescott (HP), Christiano-Fitzgerald (CF) and Baxter-King (BK) filters. According to our calculations, HP outperforms the other two filters for the pre-crisis period but offers poor end-sample estimates as expected. During the economic crisis, all models indicate a slowdown of the potential growth in Romania. The most reliable estimate for 2011 is offered by CF band-pass filter. BK filter performs poorly in the sample period with the exception of the estimate for 2008 and does not provide.