★时空地理加权回归space-time GWR GTWR分析软件-香港中文大学-黄波教授

最近在学习这方面的知识，不知道有没有用

顶一个

貌似有个STARS可以时空分析，这个工作有多位华人学者参与，其中有个牛人准备12月回武汉大学做讲座，给我回信了，他的邮箱xinyue.ye@gmail.com，可以联系一下

收藏了

非常感谢！正需要用！

謝謝樓主的分享

请问你知道怎么做时空地理加权回归模型吗？我有十年的区域经济数据

好东西，我先收下了，谢谢

gwr_st.m文件不全吧，其代码如下：
function result = gwr_st(y,x,east,north,t,info);
% PURPOSE: compute spatial and temporal geographically weighted regression
%----------------------------------------------------
% USAGE: results = gwr_st(y,x,east,north,info)
% where:   y = dependent variable vector
%          x = explanatory variable matrix
%       east = x-coordinates in space
%      north = y-coordinates in space
%       t= time coordinates
%       info = a structure variable with fields:
%       info.bwidth = scalar bandwidth to use or zero
%                     for cross-validation estimation (default)
%       info.bmin   = minimum bandwidth to use in CV search
%       info.bmax   = maximum bandwidth to use in CV search
%                     defaults: bmin = 0.1, bmax = 20                                             
%       info.dtype  = 'gaussian'    for Gaussian weighting (default)
%                   = 'exponential' for exponential weighting
%                   = 'tricube'     for tri-cube weighting
%       info.q      = q-nearest neighbors to use for tri-cube weights
%                     (default: CV estimated)  
%       info.qmin   = minimum # of neighbors to use in CV search
%       info.qmax   = maximum # of neighbors to use in CV search
%                     defaults: qmin = nvar+2, qmax = 4*nvar     
% ---------------------------------------------------
% RETURNS: a results structure
%        results.meth  = 'gwr_st'
%        results.beta  = bhat matrix    (nobs x nvar)
%        results.tstat = t-stats matrix (nobs x nvar)
%        results.yhat  = yhat
%        results.resid = residuals
%        results.sige  = e'e/(n-dof) (nobs x 1)
%        results.nobs  = nobs
%        results.nvar  = nvars
%        results.bwidth  = bandwidth if gaussian or exponential
%        results.q       = q nearest neighbors if tri-cube
%        results.dtype   = input string for Gaussian, exponential weights
%        results.iter    = # of simplex iterations for cv
%        results.north = north (y-coordinates)
%        results.east  = east  (x-coordinates)
%        results.y     = y data vector
%---------------------------------------------------
% See also: prt,plt, prt_gwr, plt_gwr to print and plot results
%---------------------------------------------------
% References: Brunsdon, Fotheringham, Charlton (1996)
% Geographical Analysis, pp. 281-298
%---------------------------------------------------
% NOTES: uses auxiliary function scoref for cross-validation
%---------------------------------------------------
% written by: James P. LeSage 2/98
% University of Toledo
%---------------------------------------------------
% revised by: Bo Huang 9/2008
% The Chinese University of Hong Kong
% References: Huang Bo (2010) Geographically and Temporally Weighted Regression for Spatiotemporal modeling of House Prices.
% International Journal of Geographical Information Science, pp. 383-401.
%---------------------------------------------------
if nargin == 6 % user options
if ~isstruct(info)
    error('gwr_st: must supply the option argument as a structure variable');
else

楼主，死骗子，花了15个币，代码都不全，有啥用，晕倒。