APPLICATION OF THE ARIMA MODEL TO FORECASTING THE CONSTRUCTION MATERIAL PRICES: A CASE STUDY IN VIETNAM

The ARIMA (Auto Regressive Integrated Moving Average) model has been widely used in forecasting time series data across various fields. Introduced by Box and Jenkins in 1974, this model offers flexibility, accuracy, and reliable confidence intervals. Its user-friendly nature, requiring a small number of parameters and hypotheses, has made it a popular choice among researchers. In Vietnam, the ARIMA model has been applied in various studies, including forecasting inflation, rice production, and construction material prices. Previous studies have demonstrated the effectiveness of the ARIMA model in short-term forecasting (1-3 periods) with stable trends or seasonality. However, the model struggles with irregular fluctuations and is less accurate in long-term forecasting. To improve forecasting accuracy, extending the multivariate model with machine learning techniques has been suggested. The ARIMA model has been used in various studies, including forecasting construction material prices in Greece, Egypt, Turkey, and Malaysia. This paper aims to analyze the fluctuation of construction material prices in Hanoi city, Vietnam, and implement the ARIMA model for forecasting. The study focuses on four essential components: cement, stone, sand, and construction steel, which are crucial for most construction projects and reinforced concrete production. The ARIMA model will be used to verify its reliability in forecasting these complex tasks. This study aims to contribute to the existing literature by analyzing the fluctuation of construction material prices in Hanoi city, Vietnam, and implementing the ARIMA model for forecasting. This research provides insights into the trends and patterns of construction material prices and evaluate the effectiveness of the ARIMA model in forecasting these complex tasks. The findings of this study can inform decision-making for construction projects and policymakers, and contribute to the development of more effective forecasting models.