Unveiling spatial and temporal patterns of second home dynamics: a Bayesian spatiotemporal analysis for a Mediterranean island

ABSTRACT

The second home is an important issue for a housing market and has effects on local land use. Conventional approaches do not consider second homes varying across space and time. Hence, spatial models, such as the Besag, York and Mollie (BYM) model and its variants, offer a feasible way to model both covariates and spatial, temporal dependence. This model is applied in our study to analyse the second home incidence rate at a county level in Corsica, France, over the period 2006–16. We investigate the spatially referenced covariates and localize ‘hot spots’ of the incidence rate across Corsican counties. Further, the temporally structured component reveals that there was a gradual increase in the incidence rate over the period under study.

KEYWORDS:

• Bayesian hierarchical modelling
• LCAR model
• Corsican second home incidence
• spatio-temporal dynamics

JEL:

• C11
• C23
• R31

ACKNOWLEDGEMENTS

The authors thank the editor-in-chief of the journal and two anonymous reviewers for their suggestions and comments on earlier drafts of the manuscript.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1 As previous stated, spatial autocorrelation is detected in the data; it is natural to consider a model that can handle both spatial correlation and potential temporal correlation tractably.

Additional information

Funding

The authors gratefully acknowledge financial support from the project ‘ESTATE’ of the University of Corsica Pascal Paoli and CNRS UMR 6240 LISA.