Interaction between Development Intensity: An Evaluation of Alternative Spatial Weight Matrices

Urban Science

article

Manman Li¹  Mengying Cui²  David Levinson³ 

[1] School of Automobile, Chang’an University;College of Transportation Engineering, Chang’an University;School of Civil Engineering, The University of Sydney

关键词: spatial weight matrix;    land use intensity;    panel data;    spatial regression model;   

DOI  :  10.3390/urbansci7010022

学科分类：社会科学、人文和艺术（综合）

来源: mdpi

【 摘 要 】

This paper investigates the spatial dependency of job and worker densities for the Minneapolis–St. Paul (Twin Cities) metropolitan area using census block level data from 2002 to 2017. A spatial weight matrix is proposed, considering the statistical expression of data, referred to as the correlation matrix, which detects the variations of dependencies among spatial units in both direction and level. The superior performance of the correlation matrix is demonstrated through a series of spatial regression models to predict land use patterns, in comparison with the conventionally used adjacency matrix as well as the accessibility matrix.

【 授权许可】

CC BY