• R语言

  • 问一个命令哈 lm ns 是啥意思呢 做出来的系数代表啥？

<br />
> summary(fit2011)</p>
<p>Call:<br />
lm(formula = y ~ ns(x, 7))</p>
<p>Residuals:<br />
ALL 8 residuals are 0: no residual degrees of freedom!</p>
<p>Coefficients:<br />
              Estimate Std. Error t value Pr(>|t|)<br />
(Intercept) -3.925e-17         NA      NA       NA<br />
ns(x, 7)1    1.509e-01         NA      NA       NA<br />
ns(x, 7)2    4.656e-01         NA      NA       NA<br />
ns(x, 7)3    5.942e-01         NA      NA       NA<br />
ns(x, 7)4    8.618e-01         NA      NA       NA<br />
ns(x, 7)5    8.878e-01         NA      NA       NA<br />
ns(x, 7)6    1.031e+00         NA      NA       NA<br />
ns(x, 7)7    9.608e-01         NA      NA       NA</p>
<p>Residual standard error: NaN on 0 degrees of freedom<br />
Multiple R-squared:     1,      Adjusted R-squared:   NaN<br />
F-statistic:   NaN on 7 and 0 DF,  p-value: NA<br />
</p>

是指的是高阶多项式吗？谢谢啦

然后如何去预测值呢？用什么方法？ 就是说知道x 求出y

那个叫ns的函数我这里报错啊。。有这个函数么？

回归模型有高阶多项式：

<br />
attach(cars)<br />
lm(dist~speed+I(speed^2))<br />
?formula<br />
</p>

样本内（外）预测：

<br />
attach(cars)<br />
predict(lm(dist~speed),data.frame(speed=20))<br />
?predict.lm<br />
</p>