Predictive Modelling for Residential Construction Demands Using ElasticNet Regression
The residential construction sector is critical to economic stability and housing availability. Residential construction demands often fluctuate due to demographic, economic, social, or market condition variables. This study seeks to investigate the significance of these external variables and produ...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Buildings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-5309/15/10/1649 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The residential construction sector is critical to economic stability and housing availability. Residential construction demands often fluctuate due to demographic, economic, social, or market condition variables. This study seeks to investigate the significance of these external variables and produce a predictive model for residential construction demand using ElasticNet regression. Adopting New Zealand as a case study and leveraging datasets from Statistics New Zealand, this research identifies key demographic, economic, and market factors influencing four building categories: retirement villages, apartments, multiunit developments, and standalone houses. The research results indicate that age groups, particularly the 20−39 and 65+ age groups, and economic indicators, such as the house price index and unemployment rates, have high prediction powers. The models showed high accuracy for some categories, with R<sup>2</sup> values exceeding 0.87 for retirement villages and 0.91 for multi-units. Challenges were encountered with standalone houses and apartments due to residual variance. The research findings highlight the importance of targeted urban planning and policy adjustments to satisfy the requirements of specific age groups, address housing affordability and demographic shifts, and cater to prevailing market conditions. This research provides practical insights and guidance for urban planners, public housing agencies, residential developers, and residential contractors while offering a robust methodological framework for predictive modelling in the construction sector. |
|---|---|
| ISSN: | 2075-5309 |