plotResid.enetLTS

0th

Percentile

residuals plots from the "enetLTS" object

Produce plots for the residuals of the current model.

Keywords
regression, classification
Usage
plotResid.enetLTS(object,vers=c("reweighted","raw"), ...)
Arguments
object

the model fit to be plotted.

vers

a character string denoting which model to use for the plots. Possible values are "reweighted" (the default) for plots from the reweighted fit, and "raw" for plots from the raw fit.

additional arguments from the enetLTS object if needed.

Value

An object of class "ggplot" (see ggplot).

Note

gives the plot of - residuals vs indices. (for both family="binomial" and family="gaussian").

- additionally, residuals vs fitted values (for only family="gaussian").

References

Kurnaz, F.S., Hoffmann, I. and Filzmoser, P. (2017) Robust and sparse estimation methods for high dimensional linear and logistic regression. Chemometrics and Intelligent Laboratory Systems.

See Also

ggplot, enetLTS, predict.enetLTS, residuals.enetLTS, fitted.enetLTS

Aliases
  • plotResid.enetLTS
Examples
# NOT RUN {
## for gaussian

set.seed(86)
n <- 100; p <- 25                             # number of observations and variables
beta <- rep(0,p); beta[1:6] <- 1              # 10% nonzero coefficients
sigma <- 0.5                                  # controls signal-to-noise ratio
x <- matrix(rnorm(n*p, sigma),nrow=n)
e <- rnorm(n,0,1)                             # error terms
eps <- 0.1                                    # contamination level
m <- ceiling(eps*n)                           # observations to be contaminated
eout <- e; eout[1:m] <- eout[1:m] + 10        # vertical outliers
yout <- c(x %*% beta + sigma * eout)          # response
xout <- x; xout[1:m,] <- xout[1:m,] + 10      # bad leverage points

# }
# NOT RUN {
fit1 <- enetLTS(xout,yout,alphas=0.5,lambdas=0.05,plot=FALSE)
plotResid.enetLTS(fit1)
plotResid.enetLTS(fit1,vers="raw")
# }
# NOT RUN {
## for binomial

eps <-0.05                                     # %10 contamination to only class 0
m <- ceiling(eps*n)
y <- sample(0:1,n,replace=TRUE)
xout <- x
xout[y==0,][1:m,] <- xout[1:m,] + 10;          # class 0
yout <- y                                      # wrong classification for vertical outliers

# }
# NOT RUN {
fit2 <- enetLTS(xout,yout,family="binomial",alphas=0.5,lambdas=0.05,plot=FALSE)
plotResid.enetLTS(fit2)
plotResid.enetLTS(fit2,vers="raw")
# }
Documentation reproduced from package enetLTS, version 0.1.0, License: GPL (>= 3)

Community examples

Looks like there are no examples yet.