Leveraging sparse Gaussian processes for property price modelling and sustainable urban planning

Abstract

This paper introduces Sparse Gaussian Processes (SGP) as an efficient solution to the computational limitations of traditional Gaussian Process Regression (GPR) in large datasets, crucial for modeling property prices. By incorporating a smaller set of inducing variables, SGPs reduce computational complexity from to and minimize storage needs, making them practical for extensive real-world applications. We apply SGPs to model property prices in Brunei, focusing on scenario analysis to evaluate different urban planning strategies’ impacts on property values. This approach aids in informed decision-making for sustainable urban development, aligning with the United Nations Sustainable Development Goal 11 (SDG 11) to foster inclusive, safe, resilient, and sustainable cities. Our findings underscore the potential of SGPs in spatial data analysis, providing a foundation for policymakers to integrate economic and environmental considerations into urban planning.

Publication
[Manuscript submitted for publication]
Haziq Jamil
Haziq Jamil
Assistant Professor in Statistics

My research interests include statistical theory, methods and computation, with applications towards the social sciences.