dc.description.abstract |
The aim of this study was to develop an Artificial Neural Network (ANN) Model for
predicting the Ghanaian Cedi to US Dollar rate and the Ghanaian Cedi to Great Britain
Pound rate with inflation, nominal growth, monetary policy, interest rate, trade balance,
gross international reserve, foreign currency deposit, broad money and US inflation. Three
different ANN models: Back Propagation Neural Network (BPNN), Radial Basis Function
Neural Network (RBFNN) and Generalized Regression Neural Network (GRNN) were
employed for the study. The results were measured by the Performance Index (PI), Mean
Absolute Error (MAE), Root Mean Square Error (RMSE) and Mean Absolute Percentage
Error (MAPE). For the Ghana Cedi - US Dollar rate, after careful and extensive training,
validation and testing, the BPNN model was realised to be the adequate model for
predicting the exchange rate with MAE of 0.28973, RMSE of 0.32274, PI of 0.10416,
MAPE of 7% and a prediction accuracy (R2) of 0.8460 as against the RBFNN which have
MAE o f0.37265, RMSE of 0.48472, PI of 0.2349, MAPE of 8.52% and an R2
of 0.3744, and the GRNN with MAE of 1.06482, RMSE of 1.15444, PI of 1.33274, MAPE of 24.07% and an R2 of 0.2987 respectively. Also, BPNN model was at the same time
identified as the sufficient model to predict the Ghana Cedi to Great Britain Pound rate
with MAE of 0.31016, RMSE of 0.38542, PI of 0.14855, MAPE of 5.62% and a
prediction accuracy (R2) of 0.7912 as against the RBFNN which have MAE of 0.40857,
RMSE of 0.51858, PI of 0.26893, MAPE of 7.02% and an R2
of 0.3705, and the GRNN with MAE of 0.50627, RMSE of 0.73248, PI of 0.53653, MAPE of 8.48% and an R2 of 0.2189 Respectively. |
en_US |