Evaluating GARCH Models for Forecasting Construction Materials Price Volatility: A Collaborative Perspective

Abstract

The volatility of construction material prices poses significant challenges for cost management and project performance in the construction industry. This study evaluates the effectiveness of Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models—specifically GARCH(1,1), EGARCH, and TGARCH—in forecasting price volatility for essential materials like cement, steel, and aggregates in Nigeria. Analyzing historical price data from January 2010 to December 2023, the research employs various model evaluation criteria, including goodness-of-fit measures (AIC, BIC) and forecast accuracy metrics (Mean Squared Error, Mean Absolute Error). The findings reveal that all GARCH models effectively capture the volatility patterns inherent in construction material prices, with the EGARCH model outperforming others in terms of both fit and predictive accuracy. The results indicate that incorporating asymmetric responses to price shocks significantly enhances forecasting capability, highlighting the importance of such considerations in volatility modeling. Accurate price forecasting is crucial for effective financial planning and risk mitigation in construction projects. The implications of this study underscore the potential of GARCH models to improve decision-making processes in the construction sector, thereby aiding quantity surveyors in managing budgets and timelines more effectively. Future research should explore the inclusion of external economic indicators to further refine forecasting precision and enhance the applicability of these models in diverse contexts within the construction industry.