Raju Rimal
16 Nov 2014
I have tried to use fortify function in ggplot2 which can access different statistics related to linear model. The basic diagnostic plot which we often get using plot function in the fitted model using lm command.
The function diagPlots gives an list of six different plots which can be arranged in a grid using grid and gridExtra packages.
require(ggplot2)
diagPlot<-function(model){
p1<-ggplot(model, aes(.fitted, .resid))+geom_point()
p1<-p1+stat_smooth(method="loess")+geom_hline(yintercept=0, col="red", linetype="dashed")
p1<-p1+xlab("Fitted values")+ylab("Residuals")
p1<-p1+ggtitle("Residual vs Fitted Plot")+theme_bw()
p2<-ggplot(model, aes(qqnorm(.stdresid)[[1]], .stdresid))+geom_point(na.rm = TRUE)
p2<-p2+geom_abline(aes(qqline(.stdresid)))+xlab("Theoretical Quantiles")+ylab("Standardized Residuals")
p2<-p2+ggtitle("Normal Q-Q")+theme_bw()
p3<-ggplot(model, aes(.fitted, sqrt(abs(.stdresid))))+geom_point(na.rm=TRUE)
p3<-p3+stat_smooth(method="loess", na.rm = TRUE)+xlab("Fitted Value")
p3<-p3+ylab(expression(sqrt("|Standardized residuals|")))
p3<-p3+ggtitle("Scale-Location")+theme_bw()
p4<-ggplot(model, aes(seq_along(.cooksd), .cooksd))+geom_bar(stat="identity", position="identity")
p4<-p4+xlab("Obs. Number")+ylab("Cook's distance")
p4<-p4+ggtitle("Cook's distance")+theme_bw()
p5<-ggplot(model, aes(.hat, .stdresid))+geom_point(aes(size=.cooksd), na.rm=TRUE)
p5<-p5+stat_smooth(method="loess", na.rm=TRUE)
p5<-p5+xlab("Leverage")+ylab("Standardized Residuals")
p5<-p5+ggtitle("Residual vs Leverage Plot")
p5<-p5+scale_size_continuous("Cook's Distance", range=c(1,5))
p5<-p5+theme_bw()+theme(legend.position="bottom")
p6<-ggplot(model, aes(.hat, .cooksd))+geom_point(na.rm=TRUE)+stat_smooth(method="loess", na.rm=TRUE)
p6<-p6+xlab("Leverage hii")+ylab("Cook's Distance")
p6<-p6+ggtitle("Cook's dist vs Leverage hii/(1-hii)")
p6<-p6+geom_abline(slope=seq(0,3,0.5), color="gray", linetype="dashed")
p6<-p6+theme_bw()
return(list(rvfPlot=p1, qqPlot=p2, sclLocPlot=p3, cdPlot=p4, rvlevPlot=p5, cvlPlot=p6))
}Using the mtcars datasets, a linear model is fitted with mpg as response and cyl, disp, hp, drat and wt has predictor variable
lm.model<-lm(mpg~cyl+disp+hp+drat+wt, data=mtcars)
diagPlts<-diagPlot(lm.model)To display the plots in a grid, some packages mentioned above should be installed.
lbry<-c("grid", "gridExtra")
lapply(lbry, require, character.only=TRUE, warn.conflicts = FALSE, quietly = TRUE)Thus the plot obtained is,
do.call(grid.arrange, c(diagPlts, main="Diagnostic Plots", ncol=3))