伍sir啊,统计是俺家不记事吃饭的家伙,他可不能忘。。。 n,p \~Tu,
J& D0,cuk
三次方的模型其实 fit 的非常好,p 值很小很小,estimates 的值如下: Nu><r
j$k/oQ
Call: kP&Ekjt@
lm(formula = V3 ~ V1 + I(V1^2) + I(V1^3), data = a) fh&Q(:ZU
C1-Jj_XQ.
Residuals: rWp+kV[Ec>
Min 1Q Median 3Q Max jVSU]LU E
-14085.3 -7779.8 -689.7 7941.2 16891.5 hM8FN
|W SvAM3
Coefficients: zH
*7!)8
Estimate Std. Error t value Pr(>|t|) KPa@~rU
(Intercept) -8.367e+05 9.471e+04 -8.835 3.07e-13 *** Xs)?PE[
V1 8.584e+03 6.063e+02 14.159 < 2e-16 *** 1+ V<-I@{
I(V1^2) -1.945e+01 1.256e+00 -15.491 < 2e-16 *** k[1w] l8
I(V1^3) 1.267e-02 8.436e-04 15.018 < 2e-16 *** &Z+.FTo
--- 9n3. Ar
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 sb%l N
hNF, sA
Residual standard error: 8724 on 75 degrees of freedom K3`48,`?wA
Multiple R-squared: 0.9952, Adjusted R-squared: 0.995 >NA{* *$0
F-statistic: 5175 on 3 and 75 DF, p-value: < 2.2e-16