A spatio-temporal analysis of house prices in Brunei Darussalam

Abstract

This study presents a novel spatio-temporal analysis of property prices in Brunei Darussalam, leveraging an extensive dataset of 3,763 residential property transactions from 2015 to 2023. This research utilises Conditionally Autoregressive (CAR) models to account for the intricate spatial and temporal dependencies observed within the housing market, a method not conventionally applied in the existing literature. Another novel contribution is in showcasing the use of the Moran’s test for testing global spatial and temporal autocorrelations simultaneously, offering a simplified approach compared to the multiple panel-based approach. The study uncovered significant spatial autocorrelation and temporal trends in property prices, allowing the coefficients to be estimated reliably while controlling for spatio-temporal heterogeneity in the data. The findings demonstrate the critical influence of spatial and temporal factors on property valuations, with an Autoregressive (AR) model of order 2 emerging as the most fitting, capturing both spatial heterogeneity and temporal persistence. This insight reveals that the housing market’s reaction to changes or shocks can linger for up to six months, underscoring the importance of considering these dimensions in real estate analysis and policy formulation.

Publication
Manuscript in submission
Haziq Jamil
Haziq Jamil
Assistant Professor in Statistics

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