BRIDGING CLASSICAL ECONOMETRICS AND MACHINE LEARNING: A STATISTICAL STRATEGY FOR NIGERIA’S PREDICTIVE AI IN  ECONOMIC PLANNING

T. A. Olominu; T. E. Balogun; I K. Adebayo; A. J. Adefioye

T. A. Olominu Academic Planning Department, Ekiti State Polytechnic, Isan Ekiti
T. E. Balogun Information Systems Department, Federal University of Technology, Akure
I K. Adebayo Department of Port Management, Nigerian Maritime University,Delta State
A. J. Adefioye Department of Statistics, Ekiti State Polytechnic, Isan Ekiti

Abstract

Nigeria’s economic planning faces growing complexity amid digital transformation, climate
volatility, and post pandemic recovery. Classical econometrics valued for its interpretability has
long guided policy, yet its limitations in handling nonlinear data constrain forecasting precision.
This study proposes a hybrid statistical framework that integrates classical econometric models
with machine learning (ML) techniques to improve predictive analytics in Nigeria's
macroeconomic management. Using a Vector Autoregression (VAR) benchmark and data-driven
ML algorithms (Random Forest, Gradient Boosting, and LSTM), the research empirically
demonstrates the forecasting advantage of hybrid models. The approach not only balances
accuracy and transparency but aligns with Nigeria’s broader vision of AI-driven policymaking and
hybrid approach not only enhances prediction accuracy across key indicators but retains the
interpretability essential for evidence-based policymaking. As Nigeria advances its National AI
Strategy and expands digital infrastructure, integrating statistical and computational methods
becomes a strategic imperative. Empowering institutions with hybrid analytics and training
practitioners in both domains will be critical to transforming data into intelligent, adaptive policy
solutions.
Keywords : Digital transformation; Classical econometrics; Machine learning (ML); Vector Auto
regression (VAR); AI-driven policymaking; Digital modernization

Author Biographies

T. A. Olominu, Academic Planning Department, Ekiti State Polytechnic, Isan Ekiti

Academic Planning Department, Ekiti State Polytechnic, Isan Ekiti

T. E. Balogun, Information Systems Department, Federal University of Technology, Akure

Information Systems Department, Federal University of Technology, Akure

I K. Adebayo, Department of Port Management, Nigerian Maritime University,Delta State

Department of Port Management, Nigerian Maritime University,Delta State

A. J. Adefioye, Department of Statistics, Ekiti State Polytechnic, Isan Ekiti

Department of Statistics, Ekiti State Polytechnic, Isan Ekiti