Today

강의

R프로그램 시즌 2 10강

결과

스터디 내용

10. 모델 평가 residuals

mcheck <- function (obj, ...){
  rs <- obj$resid
  fv <- obj$fitted
  dev.new(width = 7, height=4)
  par(mfrow=c(1,2))
  plot(fv,rs,xlab="Fitted values",ylab="Residuals",pch=16,col="red")
  abline(h=0, lty=2)
  qqnorm(rs,xlab="Normal scores",ylab="Ordered residuals",main="",pch=16)
  qqline(rs,lty=2,col="green")
  par(mfrow=c(1,1))
  invisible(NULL) }

#### normal
x <- 0:300
e <- rnorm(31,mean=0,sd=5)
yn <- 10+x+e
mn <- lm(yn~x)
mcheck(mn)

#### uniform
eu <- 20*(runif(31)-0.5)
yu <- 10+x+eu
mu <- lm(yu~x)
mcheck(mu)


#### binomial
enb <- rnbinom(31,2,.3)
ynb <- 10+x+enb
mnb <- lm(ynb~x)
mcheck(mnb)

#### gamma
eg <- rgamma(31,1,1/x)
yg <- 10+x+eg
mg <- lm(yg~x)
mcheck(mg)

Tomorrow

• R 11강 실습

Summary

• 분포별 residual & qqplot 확인