Abstract
ABSTRACT The paper proposes a Bayesian multinomial logit model to analyse spatial patterns of urban expansion. The specification assumes that the log-odds of each class follow a spatial autoregressive process. Using recent advances in Bayesian computing, our model allows for a computationally efficient treatment of the spatial multinomial logit model. This allows us to assess spillovers between regions and across land-use classes. In a series of Monte Carlo studies, we benchmark our model against other competing specifications. The paper also showcases the performance of the proposed specification using European regional data. Our results indicate that spatial dependence plays a key role in the land-sealing process of cropland and grassland. Moreover, we uncover land-sealing spillovers across multiple classes of arable land.
Highlights
Increased urbanisation and expansion of cities as a direct result of economic and population growth, coupled with intensifying climate change, poses a key challenge for policy makers (IPBES, 2019)
In this paper we put forth a Bayesian estimation approach for a multinomial logit specification for the modelling of land use conversion, which has a spatial autoregressive structure in the log odds, with differing strength of spatial autocorrelation for each choice alternative
The virtue of our specification is that it combines a spatial autoregressive framework, and a joint multinomial framework
Summary
Increased urbanisation and expansion of cities as a direct result of economic and population growth, coupled with intensifying climate change, poses a key challenge for policy makers (IPBES, 2019). Within a regional econometric framework, the land use expansion decision can be modeled as a random choice, with the multinomial logit model representing a popular option (Lubowski et al, 2008; Chakir, 2009). While the log-linearized version of the model represents a popular choice due to its ease of transformation, it suffers from the usual problems of log-transformation, namely that frequently land use shares are zero and accommodating these observations inherently biases the estimates Spatial dependence, both from unobserved spatially varying variables, as well as contingent on the choice of neighbouring regions, is well documented in the land use choice literature (Chakir and Parent, 2009; Chakir and Le Gallo, 2013; Li et al, 2013). First and foremost we present a novel Bayesian approach for capturing spatial dependence among land use changes using a multinomial logit framework.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
