Statistical Analysis Of Spatial Variable Coefficient Model

Spatial varying coefficient model is an extension of the varying coefficient model, it aims to analyze the spatial features of the data by study the model coefficient. This model has widely used in many fields,such as, finance, environmental science, ecology and epidemiology. because of it has significant effect on analysis spatial heterogeneity and the coefficient functions have a strong soul. In recent years, the research on theoretical have made great progress, while it is mainly concentrated on the application. This paper mainly studies the method of spatial varying coefficient model when its coefficient is multivariate function, at the same time discussed the coefficient functions estimate method of generalized spatial varying coefficient models and the test method of model.Firstly, studied the estimate of model coefficient function when the varying coefficient model function in the space of the functional coefficient model for multivariate function, at the same time give a new estimation model, and expounded the feasibility and effectiveness of this kind of estimate in theory. Then, discussed the influence on the model estimation base on the censored data model.Secondly, analysis of the general spatial varying coefficient model estimation and testing and introduces the common method of estimation. At the same time, studied the generalized spatial varying coefficient models’estimation and testing when it coefficient function has two variables,and discussed the local likelihood estimator and test statistics in theory.Finally, given the application of spatial varying coefficient models when coefficient function has two variables through the example analysis, further illustrate the advantages of the spatial varying coefficient model on solving the spatial features.