Construction Time Series Forecasting Using Univariate Time Series Models

Abstract

Construction costs generally change over time. Therefore, forecasting construction costs is essential for effective cost management of construction projects. This chapter aims to introduce several construction time series variables, such as Highway Construction Spending and National Highway Construction Cost Index, and demonstrate procedures for investigating the characteristics of such time series. This chapter introduces several univariate time series forecasting models, such as moving average, autoregressive, and Holt-Winters exponential smoothing models. It also provides practical examples for training and using them for forecasting construction time series. The chapter also includes R Codes, solutions, model diagnostics, and interpretations of the solutions.