星巴克在中国大陆的空间扩散特征与影响因素研究

曾国军, 陆汝瑞

Spatial expansion mode and its influencing factors ofStarbucks in mainland of China

Guojun ZENG, Rurui LU

表7 基于Matlab 7.0的nlinfit非线性最小二乘法源程序

Tab. 7 NLS source of the nlinfit based on the MATLAB 7.0

load SJ x=A(:,1:2); y=A(:,1); beta0=[-1 2.214 -0.089]; %初始值 f=@(beta,x)(beta(1)*x(:,1)+beta(1)*(x(:,1).^2))./(x(:,2).^beta(3)); [beta,resid,J,Sigma]= nlinfit(x,y,f,beta0); beta %回归系数 ci = nlparci(beta,resid,'covar',Sigma) ci %回归系数的置信区间 [ypred, delta] = nlpredci(f,x,beta,resid,'Covar',Sigma); ypred %预测值 R2=1-sum((y-ypred).^2)/sum((y-mean(y)).^2) % R2判决系数 plot([1:32],y,[1:32],ypred)