R语言一元线性回归

X <- matrix(c(
194.5, 20.79, 1.3179, 131.79,
194.3, 20.79, 1.3179, 131.79,
197.9, 22.40, 1.3502, 135.02,
198.4, 22.67, 1.3555, 135.55,
199.4, 23.15, 1.3646, 136.46,
199.9, 23.35, 1.3683, 136.83,
200.9, 23.89, 1.3782, 137.82,
201.1, 23.99, 1.3800, 138.00,
201.4, 24.02, 1.3806, 138.06,
201.3, 24.01, 1.3805, 138.05,
203.6, 25.14, 1.4004, 140.04,
204.6, 26.57, 1.4244, 142.44,
209.5, 28.49, 1.4547, 145.47,
208.6, 27.76, 1.4434, 144.34,
210.7, 29.04, 1.4630, 146.30,
211.9, 29.88, 1.4754, 147.54,
212.2, 30.06, 1.4780, 147.80),
ncol=4, byrow=T,
dimnames = list(1:17, c("F", "h", "log", "log100")))
forbes<-data.frame(X)
x<-forbes$F
y<-forbes$log100
plot(x, y)
lm.sol <- lm(y ~ 1+x)
summary(lm.sol)
abline(lm.sol)#将得到直线方程画在散点图上

#利用residuals()函数计算回归方程的残差
y.res<-residuals(lm.sol);
plot(y.res)
text(12,y.res[12], labels="残差异常点",adj=1.1)

#剔除残差异常点
i<-1:17; 
forbes12<-as.data.frame(X[i!=12, ])
lm12<-lm(log100~F, data=forbes12)
summary(lm12)