car (version 3.0-12)

infIndexPlot: Influence Index Plot

Description

Provides index plots of influence and related diagnostics for a regression model.

Usage

infIndexPlot(model, ...)
influenceIndexPlot(model, ...)
# S3 method for lm
infIndexPlot(model, vars=c("Cook", "Studentized", "Bonf", "hat"),
    id=TRUE, grid=TRUE, main="Diagnostic Plots", ...)
# S3 method for influence.merMod
infIndexPlot(model,
    vars = c("dfbeta", "dfbetas", "var.cov.comps",
    "cookd"), id = TRUE, grid = TRUE, main = "Diagnostic Plots", ...)
# S3 method for influence.lme
infIndexPlot(model,
    vars = c("dfbeta", "dfbetas", "var.cov.comps",
    "cookd"), id = TRUE, grid = TRUE, main = "Diagnostic Plots", ...)

Arguments

model

A regression object of class lm, glm, or lmerMod, or an influence object for a lmer, glmer, or lme object (see influence.mixed.models). The "lmerMod" method calls the "lm" method and can take the same arguments.

vars

All the quantities listed in this argument are plotted. Use "Cook" for Cook's distances, "Studentized" for Studentized residuals, "Bonf" for Bonferroni p-values for an outlier test, and and "hat" for hat-values (or leverages) for a linear or generalized linear model, or "dfbeta", "dfbetas", "var.cov.comps", and "cookd" for an influence object derived from a mixed model. Capitalization is optional. All but "dfbeta" and "dfbetas" may be abbreviated by the first one or more letters.

main

main title for graph

id

a list of named values controlling point labelling. The default, TRUE, is equivalent to id=list(method="y", n=2, cex=1, col=carPalette()[1], location="lr"); FALSE suppresses point labelling. See showLabels for details.

grid

If TRUE, the default, a light-gray background grid is put on the graph.

…

Arguments passed to plot

Value

Used for its side effect of producing a graph. Produces index plots of diagnostic quantities.

References

Cook, R. D. and Weisberg, S. (1999) Applied Regression, Including Computing and Graphics. Wiley.

Fox, J. (2016) Applied Regression Analysis and Generalized Linear Models, Third Edition. Sage. Fox, J. and Weisberg, S. (2019) An R Companion to Applied Regression, Third Edition, Sage.

Weisberg, S. (2014) Applied Linear Regression, Fourth Edition, Wiley.

See Also

cooks.distance, rstudent, outlierTest, hatvalues, influence.mixed.models.

Examples

# NOT RUN {
influenceIndexPlot(lm(prestige ~ income + education + type, Duncan))

# }
# NOT RUN {
 # a little slow
  if (require(lme4)){
      print(fm1 <- lmer(Reaction ~ Days + (Days | Subject),
          sleepstudy)) # from ?lmer
      infIndexPlot(influence(fm1, "Subject"))
      infIndexPlot(influence(fm1))
      }
      
  if (require(lme4)){
      gm1 <- glmer(cbind(incidence, size - incidence) ~ period + (1 | herd),
          data = cbpp, family = binomial) # from ?glmer
      infIndexPlot(influence(gm1, "herd", maxfun=100))
      infIndexPlot(influence(gm1, maxfun=100))
      gm1.11 <- update(gm1, subset = herd != 11) # check deleting herd 11
      compareCoefs(gm1, gm1.11)
      }
    
# }