Spatial stochastic regression modelling of urban land use

8th International Symposium of the Digital Earth

地球科学;计算机科学

Arshad, S.H.M.^1 ; Jaafar, J.^1 ; Abiden, M.Z.Z.^1 ; Latif, Z.A.^1 ; Rasam, A.R.A.^1

Centre of Studies for Surveying Science and Geomatics, Faculty of Architecture, Planning and Surveying, Universiti Teknologi MARA (UiTM), 40450, Shah Alam, Malaysia^1

关键词: Development strategies;    Geographically weighted regression;    Ordinary least squares;    Regression modelling;    Spatial stochastic;    Spatial structure models;    Standard of living;    Urban growth modeling;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/18/1/012170/pdf
DOI  :  10.1088/1755-1315/18/1/012170

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Urbanization is very closely linked to industrialization, commercialization or overall economic growth and development. This results in innumerable benefits of the quantity and quality of the urban environment and lifestyle but on the other hand contributes to unbounded development, urban sprawl, overcrowding and decreasing standard of living. Regulation and observation of urban development activities is crucial. The understanding of urban systems that promotes urban growth are also essential for the purpose of policy making, formulating development strategies as well as development plan preparation. This study aims to compare two different stochastic regression modeling techniques for spatial structure models of urban growth in the same specific study area. Both techniques will utilize the same datasets and their results will be analyzed. The work starts by producing an urban growth model by using stochastic regression modeling techniques namely the Ordinary Least Square (OLS) and Geographically Weighted Regression (GWR). The two techniques are compared to and it is found that, GWR seems to be a more significant stochastic regression model compared to OLS, it gives a smaller AICc (Akaike's Information Corrected Criterion) value and its output is more spatially explainable.