engleski

Confidence Indicators in Forecasting the Probability of Expansion - Evidence from Croatia

Business and Consumer Surveys (BCS) are widely known sources of information for a variety of macroeconomic analysis. It is well known that Confidence Indicators derived from business surveys are the leading indicators for the referent macroeconomic time series. The aim of this paper is, using probability models, to investigate how reliable are the Confidence Indicators in forecasting the probability of expansion. In the study, we considered three Croatian Business Survey indicators: Industrial Confidence Indicator (ICI), Construction Confidence Indicator (BCI) and Retail Trade Confidence Indicator (RTCI). These indicators are calculated in accordance with the harmonized European Union methodology. The empirical analysis consists of two parts. In the first part, for distinguishing the periods of expansion from the periods of recession in the Croatian economy, the non-parametric Bry-Boschan algorithm based on the real seasonally adjusted quarterly GDP data is used. In the second part, various probit models were estimated. Models differ in the regressors (Confidence Indicators) and in their time lags. Positive signs of estimated parameters suggest that the probability of the expansion increases with an increase in Confidence Indicators. Based on the obtained results, it can be concluded that ICI (compared to RTCI and BCI) is the most powerful predictor of expansion probability in Croatia. The quarterly data, used in the research, covered the period from 1999Q1 to 2014Q1. The data sources were Privredni vjesnik and Eurostat.

confidence indicators; expansion; business survey; probability models

Rad je u potpunosti financiran sredstvima Hrvatske zaklade za znanost, projekt IP-2013-3858