Abstract
Generalized Linear Models (GLMs) serve as a powerful extension of traditional linear regression, enhancing statistical analysis capabilities. This study employs a GLM to forecast the Return on Investment (ROI) within Nigeria’s real estate and construction sectors, utilizing a comprehensive dataset that includes data from the Nigerian Stock Exchange (NSE) and macroeconomic indicators from the National Bureau of Statistics (NBS) and the Central Bank of Nigeria (CBN), spanning 2017 to 2023. The model integrates critical macroeconomic indicators, including Gross Domestic Product (GDP) growth, the Consumer Price Index (CPI), Interest Rates (IR), and the Unemployment Rate (UR), while also considering interaction effects and non-linear terms to enhance predictive accuracy. Two methods were used to assess the GLM: Genetic Algorithms (GLM-GA) and Iteratively Reweighted Least Squares (GLM-IRLS), both of which revealed significant insights. Adjusted R2 values ranged from 0.68 to 0.73, with the highest in 2019 (0.73) and the lowest in 2021 (0.68). The Bayesian Information Criterion (BIC) exhibited variation, with values between 365.50 in 2019 and 386.30 in 2021, indicating differing model efficiency across years. Investment risk metrics, such as Value at Risk (VaR) and Conditional Value at Risk (CVaR), showed upward trends, with VaR increasing from 7.50 in 2017 to 9.20 in 2022, and CVaR rising from 7.00 to 8.70 in the same period, reflecting heightened risk exposure. The findings underscore the sensitivity of ROI predictions to macroeconomic conditions and highlight the challenges of forecasting amid economic volatility. This study emphasizes the substantial impact of macroeconomic factors on ROI and the necessity of considering these variables in investment risk assessments.