Construction Price Prediction Using Vector Error Correction Models

Abstract

Reliable prediction of construction prices is essential for the construction industry because price variation can affect the decisions of construction contractors, property investors, and related financial institutions. Various modeling and prediction techniques for construction prices have been studied, but few researchers have considered the impact of global economic events and the seasonality of construction prices. In this study, global economic events and construction price seasonality as intervention dummies, together with a group of macroeconomic variables, are considered in a vector error correction (VEC) model to accurately predict the movement of construction prices. The proposed prediction model is verified against a series of diagnostic statistical criteria and compared with conventional VEC, multiregression, and Box-Jenkins approaches. Results indicate that the VEC model with dummy variables is more effective and reliable for forecasting construction prices. The VEC model with dummy variables can also assist construction economists to analyze the effect of special events and factors on the construction industry.