Automatic stationary detection of time series using auto-correlation coefficients and LVQ — Neural network

Abstract

A data mining of Time Series using Autocorrelation Coefficients (ACC) and LVQ -Neural Network is addressed in this work-a problem that has not yet been seen in a signal processing framework, to the best of our knowledge. Neural network classification was performed on real Time series Data of real data, in an attempt to experimentally investigate the connection between Time Series data and hidden information about the properties of stationary Time Series. Finally, the ability of the ACC will be tested via a well fitted LVQ neural network which gives satisfactory results in predicting Time Series.