Unobserved Component Model for Forecasting Polish Inflation
This paper aims to use the local level models with GARCH and SV errors to predict Polish inflation. The series to be forecast, measured monthly, is consumer price index (CPI) in Poland during 1992-2008. We selected three forecasting models i.e. LL-GARCH(1,1) with Normal or Student errors and LL-SV. A simple AR(2)-SV model is used as a benchmark to assess the accuracy of prediction. The presented results indicate, that there is no clear advantage of LL models in forecasting Polish inflation over standard AR(2)-SV model, although all the models give satisfactory results.
When requesting a correction, please mention this item's handle: RePEc:cpn:umkdem:v:10:y:2010:p:121-129. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Miroslawa Buczynska)
If references are entirely missing, you can add them using this form.