Predicting Economic Resilience in Nigeria Using Machine Learning: A Framework for Policy Intervention

Abstract

Economic resilience is important in sustaining the state of the Nigerian economy from different distortions. Therefore, the current study implemented a machine learning method for estimating and forecasting Nigeria's economic resilience by using the following principal macroeconomic indicators: GDP growth, inflation, exchange rate, unemployment, debt-GDP ratio, and foreign reserves through Decision Trees (DT) and Random Forest (RF) algorithms. Amongst these, the prediction accuracy was approximately 86% by Random Forest in terms of predicting an economic downturn when compared to Decision Tree. Thus, GDP growth, inflation, and exchange rate variability were singled out as the key predictors of economic resilience. This implies that the policy ramifications of the machine learning model results are geared toward controlling inflation, stabilizing the exchange rate, creating jobs, and promoting economic diversification. The results provide data-informed policy-making to support the resilience features of the Nigerian economy.