CMStatistics 2018: Start Registration
View Submission - CFE
Title: Inflation forecast uncertainty in the three transitional Central and East European countries Authors:  Wojciech Charemza - Vistula University (Poland)
Carlos Diaz - University of Leicester (United Kingdom)
Svetlana Makarova - University College London (United Kingdom) [presenting]
Abstract: It is argued that for some countries simple measures of uncertainty based on inflation forecast errors might be more useful than the uncertainty indices based on the external text search indices which use the simple phrase and term counting method. We consider the three post-communist countries -- Poland, Russia, and Ukraine -- and the US as a benchmark for comparison. The distribution of inflation forecast uncertainty has been modelled with the use of the weighted skew-normal (WSN) distribution. The fit of the WSN distribution has been compared with that of other frequently used distributions, like the two-piece normal and generalized beta. The Monte Carlo study has shown that it is difficult to distinguish, on the statistical grounds, between these three distributions. Our results show that the WSN provides a good fit for Poland and the US, that is, for countries with a clearly defined monetary policy, but not for Russia and Ukraine. The estimates of the distributional parameters suggest the prevalence of anti-inflationary policy over the output-stimulating policy for Poland, while the opposite seems to be true for the US.