Font Size: a A A

Regional And Dynamical Real Estate Early Warning Model Based On Spatial Statistical Analysis Of GIS

Posted on:2012-06-23Degree:DoctorType:Dissertation
Country:ChinaCandidate:J ZhaoFull Text:PDF
GTID:1119330332488788Subject:Cartography and Geographic Information Engineering
Abstract/Summary:PDF Full Text Request
In past years, the real estate industry develops rapidly. Higher house prices have become one of the important factors, which affect national economic growth steadily. So it is important to establish credible regional real estate early warning model. Purpose of it is to master present situation for the real estate market timely and accurately. Through of objective analysis, trend of development will be predicted timely. In conclusion, alert will be released immediately when risk appears. All above have important meaning for the whole industry sustainable development. This paper introduces modeling of regional real estate warning, which is based on traditional statistical analysis and GIS spatial statistical analysis theory and methods. Research approaches of modeling include linear regression model, nonlinear logistic regression model and spatial linear regression model. A system framework for regional real estate early warning information release platform is also presented, which is based on Web Services. The main research results of this paper are included:1. The multiple logistic regression early warning model and the multiple linear regression early warning model are proposed. The reginal real estate early warning index system is proposed by correlation method, and early warning interval is proposed by 3 method.We provide time series data of real estate early warning, which is from 1996 to 2008, in Jinan, Shandong province. The predict result of the multiple logistic regression early warning model and the multiple linear regression early warning model are also hotly in 2009.2. Regional real estate warning spatial linear regression model is presented. The model depends on spatial regionalism, also uses spatial linear regression theory, spatial clustering and spatial autocorrelation method. It is to archive function of spatial tendency prediction. We select 740 residential points in centre of Jinan, divide 30 early warning areas by spatial regionalism, and build regional real estate warning spatial linear regression model. In conclusion, the model show statistical significant because Wald statistical value is 0.028112.3. Web services-based real-time release information platform for dynamical early warning is realized, which is setup on the ArcGIS Server architecture, a distributed component of WebGIS. Through of case analysis, we conclude that information sharing and releasing are based on ArcGIS Server. Furthermore, spatial data collection, organization, storage, and spatial statistic analysis are based on ArcGIS Destop. So we provide an integrated spatial information services system of regional real estate warning.
Keywords/Search Tags:real estate early warning, logistic regression, spatial regression, GIS, web services
PDF Full Text Request
Related items