Forecasting of Ghana’s Interest Rates Using Artificial Neural Networks

Accurate and reliable interest rate forecasts are critical for informed decisionmaking by investors, financial institutions, and policymakers. While various models and techniques have been developed to predict future interest rate movements,
the complex and dynamic nature of interest rate trends makes it dicult to accurately predict. In recent years, there has been growing interest in the use of
artificial neural networks (ANNs) for interest rate forecasting due to their ability
to capture nonlinear relationships and patterns that may be dicult to identify
using traditional modeling techniques.
A two-layer feed-forward Neural Network architecture which uses the
Levenberg - Marquardt Back propagation Algorithm was used to forecast the
Interest Rate of Ghana with Gross Domestic Product, Inflation, Money Supply
and Unemployment Rates as the macro-economic indicators that a↵ect interest
rate.The results were measured with the Mean Squared Error (MSE), Root Mean
Squared Error (RMSE), Mean Absolute Error (MAE), Weighted Absolute Percentage Error (WAPE) and the Correlation Coecient (R). After careful and
extensive training, validation and testing, the Artificial Neural Network (ANN)
produced MSE, RMSE, MAE, WAPE an R-value of 0.00108, 0.03286, 0.300,
4.24% and 0.99297 respectively with a prediction accuracy of 95.76%.
The results of this study indicate that the ANN model outperforms existing forecasting models in terms of accuracy and reliability. The study demonstrates that ANNs can capture the complex and dynamic nature of interest rate
trends and provide more accurate and reliable predictions of future interest rate
movements. The findings of this research have significant implications for financial decision-making, risk management, and macroeconomic policy.

Ghana's Interest Rate, Macro-economic indicators, Artificial NeuralNetwork (ANN), Network Architecture, Forward Pass, Backpropagation, Performance.

1. Bank of Ghana. (n.d.). Our monetary policy framework. Bank of Ghana. [Google Scholar] [Crossref]

2. Foster, W. R., Collopy, F., & Ungar, L. H. (1992). Neural network forecasting of short, noisy time series. Computers & Chemical Engineering, 16(4), 293–297. https://doi.org/10.1016/0098-1354(92)80049-F [Google Scholar] [Crossref]

3. Hagan, M. T., & Menhaj, M. B. (1994). Training feedforward networks with the Marquardt algorithm. IEEE Transactions on Neural Networks, 5(6), 989–993. https://doi.org/10.1109/72.329697 [Google Scholar] [Crossref]

4. Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359–366. https://doi.org/10.1016/0893-6080(89)90020-8 [Google Scholar] [Crossref]

5. Hsieh, W. W. (2004). Nonlinear multivariate and time series analysis by neural network methods. Reviews of Geophysics, 42(1), RG1003. https://doi.org/10.1029/2002RG000112 [Google Scholar] [Crossref]

6. Kovanen, A. (2011). Monetary policy transmission in Ghana: Does the interest rate channel work? (IMF Working Paper No. 11/275). International Monetary Fund. https://doi.org/10.5089/9781463925314.001 [Google Scholar] [Crossref]

7. Levenberg, K. (1944). A method for the solution of certain non-linear problems in least squares. Quarterly of Applied Mathematics, 2(2), 164–168. https://doi.org/10.1090/qam/10666 [Google Scholar] [Crossref]

8. Marquardt, D. W. (1963). An algorithm for least-squares estimation of nonlinear parameters. Journal of the Society for Industrial and Applied Mathematics, 11(2), 431–441. https://doi.org/10.1137/0111030 [Google Scholar] [Crossref]

9. Nti, A. E. (2021). Artificial neural network model for predicting exchange rate in Ghana [MPhil thesis, University of Mines and Technology]. University of Mines and Technology Institutional Repository. [Google Scholar] [Crossref]

10. Nunes, M., Gerding, E., McGroarty, F., & Niranjan, M. (2019). A comparison of multitask and single task learning with artificial neural networks for yield curve forecasting. Expert Systems with Applications, 119, 362–375. https://doi.org/10.1016/j.eswa.2018.11.012 [Google Scholar] [Crossref]

11. Ogundele, O. S., Ujunwa, A., & Mohammed, A. A. (2020). A neural network scheme for monetary policy rate validation in Nigeria. American Journal of Artificial Intelligence, 4(2), 62–72. https://doi.org/10.11648/j.ajai.20200402.13 [Google Scholar] [Crossref]

12. Oh, K. J., & Han, I. (2000). Using change-point detection to support artificial neural networks for interest rates forecasting. Expert Systems with Applications, 19(2), 105–115. https://doi.org/10.1016/S0957-4174(00)00025-7 [Google Scholar] [Crossref]

13. Wegener, C., von Spreckelsen, C., Basse, T., & von Mettenheim, H.-J. (2016). Forecasting government bond yields with neural networks considering cointegration. Journal of Forecasting, 35(1), 86–92. https://doi.org/10.1002/for.2385 [Google Scholar] [Crossref]