Residential construction demand forecasting using economic indicators: a comparative study of artificial neural networks and multiple regression

Abstract

In recent years, demand for residential construction has been growing rapidly in Singapore. This paper proposes the use of economic indicators to predict demand for residential construction in Singapore. At the same time, two forecasting techniques are applied, namely, Artificial Neural Networks (ANN) and Multiple Regression (MR), the former being a state-of-the-art technique while the latter a conventional one. A comparative study is carried out to determine whether the use of economic indicators with the application of the ANN technique can produce better predictions than with the MR method. A total of 12 economic indicators are identified as significantly related to demand for residential construction. Quarterly data from these 12 indicators are used to develop the ANN model. In order to assess the forecasting performance of this state-of-the-art technique, the same set of data is used to develop a conventional MR model. A comparison is made between the two models, in terms of their forecasting accuracy, by using a relative measure known as the Mean Absolute Percentage Error (MAPE). The forecasting error of the ANN model is found to be about one fifth of that derived from the MR model. The low MAPE values (less than 10%) obtained for both models also indicate that economic indicators may be used as reliable inputs for the modelling of residential construction demand in Singapore.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 86 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.