为什么MATLAB多项式拟合，拟合次数越高，最小误差值越大呀？是因为程序错了吗 >> x=1:9

为什么MATLAB多项式拟合，拟合次数越高，最小误差值越大呀？是因为程序错了吗 >> x=1:9; y=[12.5 11.9 10.5 10 9.6 9.5 9.5 9.3 8.5]; y_hat=0.0013*x.^5-0.00395*x.^4+0.4179*x.^3-1.8151*x.^2+2.3345*x+11.6250; S=sum((y_hat-y).^2) S = 8.5161e+04 >> x=1:9; y=[12.5 11.9 10.5 10 9.6 9.5 9.5 9.3 8.5]; y_hat=-0.0067*x.^4+0.1164*x.^3-0.5777*x.^2+0.1493*x+12.8778; S=sum((y_hat-y).^2) S = 0.1425