Implementing an Intelligent Moving Average with a Neural Network

Recent results in hybrid neural networks using extended versions of the core method have shown that we can use background knowledge to guide back-propagation learning. This paper further explores this ideas by adding numeric functions to the encoded knowledge and using the traditional recursive Elman neural network model. An illustration of the properties of these neural networks will be used to calculate a simple moving average. Simulations on generated data and on the Eurostoxx50 financial index will illustrate the potential of such a strategy.