Timotej Jagrič, Sebastjan Strašek
Forecasting economic activity in a transition economy as leading indicator approach
Journal of European Economy, Volume 2, Issue 1, March 2003, pp. 115-130
|JEL: C32, E32||Full text (PDF)|
The future income level is of central importance for a transition economy. Since the level of income is strongly related to aggregate economic activity, the measurement of future activity demands appropriate forecasting model. To construct such model it is necessary to select an indicator of economic activity, as well as a group of variables that, when adjusted, construct the composite (CLI) and diffuse (DLI) leading indicators that forecast the reference series. We develop a model where NBER method is modified with elements of Stock-Watson approach. The model is applied then on a small transition economy (Slovenia). The ex-post analysis suggests that in the period from 1992 to 2001, CLI and DLI forecast all turning points of aggregate economic activity. The average lead-time was 8 years, which is comparable with the performance of leading indicators in other countries.
bond, business cycle, composite index, diffuse index, forecast, Granger causality, investor, leading indicator, level of income, price, securities, visible cycle, share, spectral analysis, stock exchange, structural changes, time series, invisible cycle, cause and effect.