Estimation Methods And Predictive Possibilities Of Core Inflation: Primjer Hrvatske

This paper analyses different core inflation estimation methods in the case of the Republic of Croatia. Namely, monitoring and controlling of inflation are extremely important because of the long run consequences they have on the economy as a whole. Hence, it is necessary to predict the future movements of inflation with certainty, in order to react in time. Many predictors of future inflation, including experts of central banks, use core inflation rate as an indicator of future trend of headline inflation. With regard to core inflation, CPI inflation will be under higher influence of temporary changes in relative prices (because food and energy prices are less stabile than other inflation components). Thereby, this paper implements calculation of core inflation which tries to remove these transitory fluctuations and price level changes by computing limited-influence estimators - the trimmed mean and the median. Furthermore, predictive possibility of core inflation measures, computed as trimmed mean and median, are based on regression models and their errors of prediction. The results of econometric analysis show that trimmed mean inflation is statistically justified, because it has the best indicators of forecast error statistics, which refer to the conclusion that creators of economic policy could improve the quality of decision making based on forecasting model of inflation.