Artificial neural network application in modeling revenue returns from mobile payment services in Kenya

摘要

Artificial Neural Networks has recently shown a great applicability in time-series analysis and forecasting thus correctly deducing the unseen part of the population even if the sample data contain noisy information. In this paper we used Neural Network to model revenue returns from mobile payment services using dataset extracted from Central Bank of Kenya website. The network with one or two hidden layers was tested with various combination of neurons, and results were compared in terms of forecasting error. It was observed that ANN if properly trained accurately forecast Revenue returns on mobile payments services in Kenya.