matlab中怎么用最小二乘法求目标函数的最优解

xi=[0.019,0.023,0.027];%xdatayi=[430,380,256];%ydataabc0=[4,0.5,5];%initialvalues%%初始值设定abc=lsqcurve

x0=[12345678];y0=[1.121.983.013.975.116.136.998.05];p=polyfit(x0,y0,1);y=poly2str(p,'x')a=p(1)b=p(2)

x1=[350350350350350350350350350500500500500500500500500500650650650650650650650650650];x2=[2223.53.5

F=@(p,x)p(1)*x(:,1).^2+p(2)*x(:,1).*x(:,2)+p(3)*x(:,2).^2+p(4);%离散数据点x=[6.324606.93791.13727.08751.4

用函数lsqnonlin进行最小二乘拟合functionf=myfunc(c,x,y)f=y-c(1)+c(2)./x;return保存为M文件.然后在命令行下输入：x=[你的x数组]y=[你的y数组

t=[11.41.622.42.633.644.656];y=[0.264241,0.408167,0.475069,0.593994,0.691559,0.732615,...0.800852,0.

1）>>x=linspace(1,32,8);>>y=linspace(0.1,0.8,8);>>p=polyfit(x,y,1)p=0.02260.0774%a=0.0226b=0.07742）>>

首先确认一下,你这的log是以10为底的对数吧?那么命令如下.（如果是以e为底就不用除以log(10)了,自己改一下.)>>x=0.01:0.01:1;>>m=log(x)/log(10);>>for

y-1=exp(x/a)^2两边取对数,化成线性拟和然后polyfit

x=[-2.573\x09-2.159\x09-1.741\x09-1.403\x09-1.029\x09-0.699];y=[0.640\x090.844\x091.081\x091.227\x09

F=@(p,x)p(1)*x(:,1).^2+p(2)*x(:,1).*x(:,2)+p(3)*x(:,2).^2+p(4);%离散数据点x=[6.324606.93791.13727.08751.4

x=[1,2,3,4,5]y=[500.6,442.4,428.6,370.1,343.1]p=polyfit(x,y,1)y6to10=polyval(p,[6:10])plot(x,y,'o',[

如上所述,采用polyfit来拟合,二次多项式polyfit(x,y,2)x=[0.110.130.190.210.270.370.530.590.710.790.891.07];y=[3868-10

clc%原数据节点x=[0.511.522.53];y=[1.752.453.814.888.6];plot(x,y,'*');holdon;%求最小二乘意义下拟合多项式的系数n=3;%待拟合多项式的

A'A要满秩才求逆解方程啊,所以不一定A'AX=A'b恒有解的.不过考虑到一般实际用到最小二乘很多事实验数据拟合,所以基本上A'A都是满秩的

t=0:5:55;y=[01.272.162.863.443.874.154.374.514.584.624.64];%你将t,y的每一个值代入方程,会得到关于a,b,c的三元一次方程组(12个方程)

用曲线拟合工具箱做,很简单的,不过x,y的值要到你的工作空间里去,如图,这里的a,b相当于你的c,n

y=[00.060.110.170.230.280.340.400.460.520.590.650.720.800.870.961.051.161.311.55]';x=[20.0218.9517.7

http://hi.baidu.com/nature_and_man/item/15bac2a61bd7229e151073a8matlab最小二乘法

命令窗口输入x=[0,0.5,1,1.5,2,2.5,3,3.5];y=[1,2.4,3.1,5.0,7,11,17,24];cftool在新的图形界面中,点data,creatdata,x与y选择相