COMPARISON OF INERTIA OF STATISTICAL METHODS FOR FORECASTING OF UNEMPLOYMENT UNDER CONDITIONS OF ECONOMIC DEPRESSION

Suitable models of unemployment rate development in Czech Republic under condition of economic depression were created and presented in this paper. Models’s were based on exponential smoothing and training of artificial neural networks. The most suitable models were exponential eventually damped model with additive seasonality and multilayer perceptron. March’s and April’s unemployment rate forecast was 7.58–7.89 % and 7.63–8.33 % for exponential smoothing respective 7.3–7.45 % and 6.62–8.22 % for multilayer perceptron. Recalculation of exponential smoothing model and retraining of ANNs is necessary at the time when it is possible obtain fresh values of the unemployment rate regarding to the present economic situation and relevant forecasting.