MODELLING CONSUMER PRICE INDEX DYNAMICS IN NIGERIA USING DECISION TREE AND RANDOM FOREST MACHINE LEARNING ALGORITHMS

U. A. IBEKWE; G. C. MESIKE; M. B. ASHOGBON

U. A. IBEKWE Department of Actuarial Science & Insurance Faculty of Management Sciences University of Lagos, Nigeria
G. C. MESIKE Department of Actuarial Science & Insurance Faculty of Management Sciences University of Lagos, Nigeria
M. B. ASHOGBON Department of Banking and Finance, Lagos State University of Science and Technology

Keywords: Consumer Price Index, Machine Learning, Decision Tree, Random Forest, OLS

Abstract

This study modeled Consumer Price Index (CPI) dynamics in Nigeria using historical CPI data from the Nigeria Bureau of Statistics for the period 2008-2022 inclusive. The study deployed two Machine Learning algorithms-Decision Tree and Random Forest, using Ordinary Least Squares (OLS) as benchmark. Model performances were contrasted regarding goodness of fit and prediction accuracy. The research revealed that the baseline model, OLS, turned out to give better results than both The Decision Tree and Random Forest models. The findings imply that although Random Forest performed better than Decision Tree, Ordinary Least Squares outperformed both of them.

Author Biographies

U. A. IBEKWE, Department of Actuarial Science & Insurance Faculty of Management Sciences University of Lagos, Nigeria

G. C. MESIKE, Department of Actuarial Science & Insurance Faculty of Management Sciences University of Lagos, Nigeria

M. B. ASHOGBON, Department of Banking and Finance, Lagos State University of Science and Technology