matlab实现Newton插值函数

Newton插值

1. 理论
   

2. 程序

Newton插值函数：
function p = Newton_fun(x,xi,yi)
n = length(xi);
f = zeros(n,n);
% 对差商表第一列赋值
for k = 1:n      
    f(k,1) = yi(k);
end
% 求差商表
for i = 2:n    
    for k = i:n
        f(k,i) = (f(k,i-1)-f(k-1,i-1))/(xi(k)-xi(k+1-i));  
    end
end
% 求插值多项式
P = 0;          
for k = 2:n
    t = 1;
    for j = 1:k-1
        t = t*(x-xi(j));
    end
    p = f(k,k)*t+p;
end
p = f(1,1)+p;
end

main函数：
xi = [0:0.1:2*pi];
yi = sin(xi);
x = xi;
for i = 1:length(x)
    y(i) = Newton_fun(x(i),xi,yi);
end
subplot(2,1,1);plot(xi,yi,'-*r');title('原函数');
subplot(2,1,2);plot(x,y,'-*b');title('Newton插值');

3. 实验结果