Exchange Rate Prediction Using a Multi Layer Feed-Forward Artificial Neural Networks
Loading...
Date
2013
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This research aimed at forecasting the Ghanaian Cedi – US Dolar rate with Treasury bill rates,
money supply, consumer price index and inflation
With the aim of exploring the efficiency of Artificial Neural Network (ANN) which is an
imitation of the Human Brain, a two (2) layer feedforward network using Levenberg –
Marquardt Backpropagation Algorithm was used for the forecast and the results were
measured by the Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and the
Weighted Absolute Percentage Error (WAPE).
After careful and extensive training, validation and testing, the ANN produced MSE, RMSE,
WAPE and an R-value of 0.0010, 0.0324, 2.30%, and 0.99634 respectively.
An Artificial Neural Network model was obtained which was compared with the traditional
multiple regression model with the ANN model producing a prediction accuracy of 97.70% as
compared to 76.79% of the Regression model.
Description
A Thesis Submitted to the Department of Mathematics, Kwame Nkrumah
University of Science and Technology in partial fulfillment of the
requirement for the degree of
Master of Philosophy Applied Mathematics