多元非参数回归模型的核估计
因变量Yi,自变量X1i,X2i,随即误差项{ui}独立同分布，模拟的模型为
Yi=10(X1i+1)(X2i+1)+ui  数据为d=[11.9694,0.0,0.0;7.4965,0.0,0.1;10.6182,0.0,0.2]

%matlab程序
n=3;
d=[11.9694,0.0,0.0;7.4965,0.0,0.1;10.6182,0.0,0.2];
y=d(:,1);
x=d(:,2:3);
m0=10*(x(:,1)+1).*(x(:,2)+1);
h=0.24;
m=zeros(n,1);
p=ones(n,1);
for i=1:n;
    w=zeros(n);
    for j=1:n;
        e=((x(j,1)-x(i,1))/h)^2+((x(j,2)-x(i,2))/h)^2;
        w(j,j)=(2/pi)*(1-e)*(e<1);
    end;
    m(i)=(p'*w*y)/(p'*w*p);%m(x)的估计量
end;
[y,x(:,1),x(:,2),m0,m]
sqrt(mean((m0-m).*(m0-m)))


请大侠帮忙解释一下循环体的部分:
for i=1:n;
    w=zeros(n);
    for j=1:n;
        e=((x(j,1)-x(i,1))/h)^2+((x(j,2)-x(i,2))/h)^2;
        w(j,j)=(2/pi)*(1-e)*(e<1);
    end;
    m(i)=(p'*w*y)/(p'*w*p);%m(x)的估计量
end;