lm.predictions.a1 <- predict(final.lm,clean.algae)
algae <- algae[-manyNAs(algae),]
clean.algae <- knnImputation(algae,k=10)
clean.algae
lm.a1 <- lm(a1~.,data=clean.algae[,1:12])
lm.a1
summary(lm.a1)
anova(lm.a1)
lm2.a1 <- update(lm.a1,.~.-season)
lm2.a1
summary(lm2.a1)
anova(lm.a1,lm2.a1)
final.lm <- step(lm.a1)
summary(final.lm)
####回归树
library(rpart)
library(DMwR)#加载这个包，才会有海藻数据集
data(algae)
algae <- algae[-manyNAs(algae),]
rt.a1 <- rpart(a1~.,data=algae[,1:12])
rt.a1
prettyTree(rt.a1)
printcp(rt.a1)
rt2.a1 <- prune(rt.a1,cp=0.08)
rt2.a1
first.tree <- rpart(a1~.,data=algae[,1:12])
snip.rpart(first.tree,c(4,7))
rt.a1 <- rpartXse(a1~.,data=algae[,1:12])
a
rt.a1
prettyTree(first.tree)
snip.rpart(first.tree)
lm.predictions.a1 <- predict(final.lm,clean.algae)
summary(lm.a1)
anova(lm.a1)
lm2.a1 <- update(lm.a1,.~.-season)
summary(lm2.a1)
anova(lm.a1,lm2.a1)
anova(lm2.a1)
#不断重复这个过程，直到没有可剔除的候选系数
final.lm <- step(lm.a1)
summary(final.lm)
library(rpart)
data(algae)
algae <- algae[-manyNAs(algae),]
rt.a1 <- rpart(a1~.,data=algae[,1:12])
rt.a1
prettyTree(rt.a1)
printcp(rt.a1)
rt2.a1 <- prune(rt.a1,cp=0.08)
rt2.a1
rt.a1 <- rpartXse(a1~.,data=algae[,1:12])
rt.a1
first.tree <- rpart(a1~.,data=algae[,1:12])
snip.rpart(first.tree,c(4,7))
my.tree <- snip.rpart(first.tree,c(4,7))
my.tree
prettyTree(first.tree)
snip.rpart(first.tree)
q()
rt.predictions.a1 <- predict(rt.a1,algae)
(mae.a1.lm <- mean(abs(lm.predictions.a1-algae[,"a1"])))
(mae.a1.rt <- mean(abs(rt.predictions.a1-algae[,"a1"])))
#均方误差
(mse.a1.lm <- mean((lm.predictions.a1-algae[,"a1"])^2))
(mse.a1.rt <- mean((rt.predictions.a1-algae[,"a1"])^2))
#平均绝对误差 NMSE
(nmse.a1.lm <- mean((lm.predictions.a1-algae[,'a1'])^2)/mean((mean(algae[,'a1'])-algae[,'a1'])^2))
(nmse.a1.rt <- mean((rt.predictions.a1-algae[,'a1'])^2)/mean((mean(algae[,'a1'])-a1gae[,'a1'])^2))
old.par <- par(mfrow=c(1,2)
plot(lm.predictions.a1,clean.algae[,"a1"],main="Linear Model",xlab="Predictions",ylab="True Values")
lm.predictions.a1
algae[,"a1"]
abline(0,1,lty=2)
plot(rt.predictions.a1,clean.algae[,"a1"],main="Regression Tree",xlab="Predictions",ylab="True Values")
abline(0,1,lty=2)
par(old.par)
algae[identify(lm.predictions.a1,clean.algae[,"a1"]),]
sensible.lm.preictions.a1 <- ifelse(lm.predictions.a1<0,0,lm.predictions.a1)
