EN
The aim of this article is to present original application wavelets to the prediction of short-term of time series. The model proposed to predict short-term time series (in particular for predicting macroeconomic indicators) is a model of copyright. The model is based on wavelet analysis, the Haar wavelet, the Daubechies wavelet and adaptive models. The Daubechies wavelets are a family of orthogonal wavelets and are characterized by a maximal number of vanishing moments for some given support. Adaptive models have been appropriately modified by the introduction of a wavelet function and combined into one predictive model. The results obtained from the study results indicate that the authorial model is an effective tool for short-term predictions. The model was applied to predict macroeconomic indicators.