A Neural Network Model for Time-Series Forecasting

The paper presents some aspects regarding the use of pattern recognition techniques and neural networks for the activity evolution diagnostication and prediction by means of a set of indicators. Starting from the indicators set there is defined a measure on the patterns set, measure representing a scalar value that characterizes the activity analyzed at each time moment. A pattern is defined by the values of the indicators set at a given time. Over the classes set obtained by means of the classification and recognition techniques is defined a relation that allows the representation of the evolution from negative evolution towards positive evolution. For the diagnostication and prediction the following tools are used: pattern recognition and multilayer perceptron. The paper also presents the REFORME software written by the authors and the results of the experiment obtained with this software for macroeconomic diagnostication and prediction during the years 2003-2010.