Data pooling for early-stage price forecasts

Many clients or building owners rely heavily on the early stage construction cost forecasts, provided by the Quantity Surveyor (Q.S.), for their investment decisions and advance financial arrangements. During the preliminary stage, the information concerning the new project is very scarce. It is very common in practice for Q.S. to use historical building cost data on which to base the forecast of the new project - typically basing the forecast on the known price of the most similar project to the new one. However, this approach not always generates the most accurate results. This paper develops an idea for early stage forecasting by using out-of-sample mean square errors to measure the forecasting accuracy. A method is presented for finding the best pooling arrangement of the available data source according to the characteristics of the new project. By making the best use of the available data pool, it can maximize the quality of the early stage cost forecast with limited project information.