PDA

查看完整版本 : 非线性回归出现的错误了,大家试过吗?


beardgh
2008-12-22, 17:07
m文件如下:
function [result1,result2 ]= nlinfitfunc(x,y,c01,t1)
global c0 t
% NLINFITFUNC Summary of this function goes here
% Detailed explanation goes here
% x=[1.5 2.5 3.5 5 7 9 11];
% y=[0.524 0.4273 0.2715 0.1522 0.0684 0.0385 0.043];
%初始赋值
c0=c01;
t=t1;
beta0=[0.3,0.000003];
result=nlinfit(x,y,@model,beta0);
%结果1
result1=result(1);
%结果2
result2=result(2);
format long e;
end
function y=model(beta0,x)
global c0
global t
%MODEL Summary of this function goes here
% Detailed explanation goes here
a=beta0(1);
b=beta0(2);
%c=sqrt(4*b*t*24*60*60);
%d=x/c;
%format long e
%y=c0+(a-c0)*(1-erf(d));
y=c0+(a-c0)*(1-erf(x/sqrt(4*b*t*24*60*60)));
end

用的时候出现这样的错误啊!!
[result1,result2 ]= nlinfitfunc(x,y,c01,t1)
??? Error using ==> erfcore
Input must be real.

Error in ==> erf at 21
y = erfcore(x,0);

Error in ==> nlinfitfunc>model at 28
y=c0+(a-c0)*(1-erf(d));

Error in ==> nlinfit>LMfit at 300
yfit = model(beta,X);

Error in ==> nlinfit at 167
[beta,J,lsiter,cause] = LMfit(X,y, model,beta,options,verbose,maxiter);

Error in ==> nlinfitfunc at 11
result=nlinfit(x,y,@model,beta0);

就是这个erf的问题