Dynamic Predictions from Time Series Data — An Artificial Neural Network Approach

Abstract

A hybrid approach, incorporating concepts of nonlinear dynamics in artificial neural networks (ANN), is proposed to model a time series generated by complex dynamic systems. We introduce well-known features used in the study of dynamic systems — time delay τ and embedding dimension d — for ANN modeling of time series. These features provide a theoretical basis for selecting the optimal size for the number of neurons in the input layer. The main outcome of the new approach for such problems is that to a large extent it defines the ANN architecture, models the time series and gives good prediction. As a consequence, we have an integrated and systematic data-driven scheme for modeling time series data. We illustrate our method by considering computer generated periodic and chaotic time series. The ANN model developed gave excellent quality of fit for the training and test sets as well as for iterative dynamic predictions for future values of the two time series. Further, computer experiments were conducted by introducing Gaussian noise of various degrees in the two time series, to simulate real world effects. We find that up to a limit introduction of noise leads to a smaller network with good generalizing capability.