The ability to predict future values of any physical or logical parameters would enable humans to plan better, exploit opportunities, improve forecasting, optimize resources and make informed decisions that can enhance the way we live.
Time series prediction is based on modelling future values of any specific parameter based on its current data. Examples of time series-based models are average rainfall forecast, internet traffic rates, business trend forecasting, weather forecast, contagious disease spread, and others.
Artificial neural networks are increasingly becoming successful in predicting time series data through regression-based models.
In this Knowledge Sharing article – awarded 2nd Place in the 2018 Knowledge Sharing Competition – Rajasekhar Nannapaneni details how the benchmark called Mackey-Glass chaotic time series model is predicted using feed forward multi-layer perceptron neural network with error back propagation algorithm. The architecture and algorithm developed in this article can easily be extended to any real world time-series prediction requirements.