Comparative analysis of methods of time series prediction based on neural networks and regression analysis

The article deals with two main areas in the prediction of time series, namely, neural network forecasting techniques and methods based on regression analysis. A comparison of the results forecast by the example of selected indicators produced by the two methods. Analyzes the main problems arising from the use of these methods, as well as methods for their solution, in particular the hybridization of these methods. Conducted a review of studies comparing the predictive performance of methods based on artificial neural networks and other methods of forecasting. Particular attention is paid to methods of comparison of ANN and multiple regression techniques.