Forecasting Architecture Billings Index Using Time Series Models: Fid-Bau Portal

Forecasting Architecture Billings Index Using Time Series Models

Kim, Sooin / Abediniangerabi, Bahram / Shahandashti, Mohsen

The construction market presents volatile fluctuations through time. These fluctuations are problematic for project planners and decision-makers to accurately assess business conditions and develop business outlooks. Therefore, it is necessary to forecast the market trends and fluctuations as accurately as possible for planning construction projects, allocating resources, and formulating strategies. Practitioners refer to Architecture Billings Index (ABI) as a leading indicator in the construction market. ABI forecasting can assist project planners in tracking construction market fluctuations to properly allocate resources and identify business opportunities. The main objective of this research is to develop time series models for forecasting ABI. The historical ABI data are used to develop the univariate time series models and validate their forecasting performance. The results show that the seasonal autoregressive integrated moving average (Seasonal ARIMA) model provides the most accurate out-of-sample forecasts. The findings of this research can assist project planners and decision-makers in forecasting volatile fluctuations in the construction market for project planning and strategic decision-making.