数值代数课程设计

0.8125 0.75 1.00 1.3125 1.75 2.3125];Р>> y=[ 1.00 0.8125 0.75 1.00 1.3125 1.75 2.3125];Р>> fy=fi-y; fy2=fy.^2; J=sum(fy.^2);Р运行后屏幕显示误差平方和如下РJ=Р(a-b+c-1)^2+(9/16*a-3/4*b+c-13/16)^2+(1/4*a-1/2*b+c-3/4)^2+(c-1)^2+(1/16*a+1/4*b+c-21/16)^2+(1/4*a+1/2*b+c-7/4)^2+(9/16*a+3/4*b+c-37/16)^2Р为求使达到最小,只需利用极值的必要条件,,,得到关于的线性方程组,这可以由下面的MATLAB程序完成,即输入程序Р>> Ja1=diff(J,a); Ja2=diff(J,b); Ja3=diff(J,c);Р>> Ja11=simple(Ja1), Ja21=simple(Ja2), Ja31=simple(Ja3) Р运行后屏幕显示J分别对的偏导数如下Р Ja11 =Р 451/128*a-63/32*b+43/8*c-887/128Р Ja21 =Р -63/32*a+43/8*b-3/2*c-61/32Р Ja31 =Р 43/8*a-3/2*b+14*c-143/8Р解线性方程组,输入下列程序Р >> A=[451/128, -63/32, -3/2 ;-63/32,43/8,-3/2;43/8,-3/2,14];Р>> B=[887/128,61/32,143/8];Р>> C=B/A, f=poly2sym(C)Р运行后屏幕显示拟合函数f及其系数C如下РC =Р 0.3081 0.8587 1.4018Рf =Р 924/2999*x^2+10301/11996*x+4204/2999Р 故所求的拟合曲线为