wine=[1.853,1.461,1.557,0.3664,0.0545,101.796 ]; 对于白葡萄的理化指标的选择，我们依据第二问中所分析出来的重要指标中选择6个重要指标：
putao=[0.2245,1.616,3.315,3.810,5.450,115.256]; wine=[0.0545,0.3664,1.461,1.557,1.853]; putao=[0.2245,1.616,3.315,3.810,5.450]; n=[1:3];
p1=polyfit(wine,putao,n(1)) p2= polyfit(wine,putao,n(2)) p3=polyfit(wine,putao,n(3)) putao1=polyval(p1,wine); putao2=polyval(p2,wine); putao3=polyval(p3,wine);
plot(wine,putao,'ko',wine,putao1,'-k*',wine,putao2,'--kx',wine,putao3,':kd');
xlabel('wine');ylabel('putao');
legend('原始数据','1次曲线','2次曲线','3次曲线'); p1 =
2.4603 0.2792 p2 =
0.8415 0.8960 0.5665 p3 =
Columns 1 through 3
2.4933 -6.6924 6.8559
Column 4
-0.1265
各次拟合曲线与原数据的比较结果如图所示，。由p3可得3次拟合曲线多项式函数为： F=p3(1)x^3+p3(2)x^2+p3(3)x+p3(4)=2.4933x^3-6.6924x^2+6.8559x-0.1265 接着求的y的3次拟合的曲线机器预测误差范围+-deltay 代码如下：
[p,s]=polyfit(wine,putao,3);
[putao3,deltay]=polyval(p,wine,s);
putaolo=putao3-deltay;putaoup=putao3+deltay;
plot(wine,putao,'ko',wine,putao2,'-k*',wine,putaolo,'-.bs',wine,putaoup,'-.bd');
xlabel('wine');ylabel('putao');
legend('原始数据','3次曲线','误差下限','误差上限')
26