Abstract
The spatial variation of geographical phenomena is a classical problem in spatial data analysis and can provide insight into underlying processes. Traditional exploratory methods mostly depend on the planar distance assumption, but many spatial phenomena are constrained to a subset of Euclidean space. In this study, we apply a method based on a hierarchical Bayesian model to analyse the spatial variation of network-constrained phenomena represented by a link attribute in conjunction with two experiments based on a simplified hypothetical network and a complex road network in Shenzhen that includes 4212 urban facility points of interest (POIs) for leisure activities. Then, the methods named local indicators of network-constrained clusters (LINCS) are applied to explore local spatial patterns in the given network space. The proposed method is designed for phenomena that are represented by attribute values of network links and is capable of removing part of random variability resulting from small-sample estimation. The effects of spatial dependence and the base distribution are also considered in the proposed method, which could be applied in the fields of urban planning and safety research.
Highlights
Spatial variation is a classic problem in spatial data analysis and can provide insight into the spatial patterns of geographic phenomena and spatial processes
The paper introduces a method based on hierarchical Bayesian models to analyse the spatial variability of network-constrained phenomena by considering the random variation resulting from small sample estimation in conjunction with two experiments based on a simplified hypothetical network and a complex road network in Shenzhen that includes 4212 urban facility points of interest (POIs) for leisure activities
The results indicate that the model, which has 50 network events with a standard basic spatial units (BSUs) length of 200, shows a higher performance with a lower deviance information criterion (DIC), and there is no considerable difference in the smoothing properties of the conditional autoregressive process (CAR) model using two types of spatial weight matrices because the differences in DIC are less than 5
Summary
Spatial variation is a classic problem in spatial data analysis and can provide insight into the spatial patterns of geographic phenomena and spatial processes. It is generally assumed that the spatial events can be located stochastically on a plane, and the spatial association between event locations or sub-areas is analysed using the Euclidean (or planar) distance [1,2,3], in which the inherent spatial processes are quantified based the assumption of Euclidean geometry [4]. This assumption is not appropriate when a spatial phenomenon is apparently constrained to a subset of geographical space, such as a street network. Network spatial analysis has been developed in the past two decades and is widely applied in analysing network-constrained spatial phenomena [7,8,9,10,11,12,13,14,15]
Sign-in/Register to view highlights & summary 
Published Version (
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