The accuracy and optimal linear correction of UK construction tender price index forecasts

This paper uses a number of quantitative methods to assess the accuracy of the UK construction tender price index forecast (TPI) produced by experts for the Building Cost Information Service (BCIS) from 1980 to 1992. The test of comparative predictive accuracy shows that the BCIS forecast is less accurate than forecasts based upon a simple (naïve) model which is widely regarded as the benchmark of satisfactory predictive performance. The study shows that the BCIS forecast contains systematic error which can be reduced with the use of optimal linear correction. The paper, apart from assessing the forecasting error of construction price movement, is normative in nature, producing aids for assessing forecast accuracy of the construction industry time series data.