Time Series Analysis of ENR Construction Cost Index

TLDR

The Engineering News‑Record publishes a monthly construction cost index that rises over time but fluctuates sharply, complicating accurate bid and budget preparation. The study aims to forecast construction cost trends so that bids and estimates can be more accurate and avoid under‑ or over‑estimation. It compares several univariate time‑series models for both in‑sample and out‑of‑sample forecasting of the index. Seasonal ARIMA performs best for in‑sample forecasts, Holt‑Winters for out‑of‑sample, and both outperform expert forecasts, allowing cost estimators, owners, and contractors to improve bids and project timing.

Abstract

Every month, Engineering News-Record (ENR) publishes the construction cost index (CCI), which is a weighted aggregate index of the 20-city average prices of construction activities. Although CCI increases over the long term, it is subject to considerable short-term variations, which make it problematic for cost estimators to prepare accurate bids for contractors or engineering estimates for owner organizations. The ability to predict construction cost trends can result in more-accurate bids and avoid under- or overestimation. This paper summarizes and compares the applicability and predictability of various univariate time series approach for in-sample and out-of-sample forecastings of CCI. It is shown that the seasonal autoregressive integrated moving-average model is the most-accurate time series approach for in-sample forecasting of CCI, while the Holt-Winters exponential smoothing model is the most-accurate time series approach for out-of-sample forecasting of CCI. It is also shown that several time series models provide more-accurate out-of-sample forecasts than the ENR’s subject matter experts’ CCI forecast. Cost estimators can benefit from CCI forecasting by incorporating predicted price variations in their estimates and preparing more-accurate bids for contractors and budgets for owners. Owners and contractors can use CCI forecasting in reducing construction costs by better-timed project execution.