plotenvelope {robustbetareg}

Normal Probability Plots of Residuals with Simulated Envelope for robustbetareg Objects

Description

plotenvelope is used to display normal probability plots of residuals with simulated envelope for the robust beta regression. Currently, eight types of residuals are supported: sweighted2, pearson, weighted, sweighted, sweighted.gamma, sweighted2.gamma, combined, and combined.projection residuals.

Usage

plotenvelope(
  object,
  type = c("sweighted2", "pearson", "weighted", "sweighted", "sweighted.gamma",
    "sweighted2.gamma", "combined", "combined.projection"),
  conf = 0.95,
  n.sim = 100,
  PrgBar = TRUE,
  control = robustbetareg.control(...),
  ...
)

Arguments

object

fitted model object of class robustbetareg.

type

character indicating the type of residuals to be used, see residuals.robustbetareg. Default is type = "sweighted2".

conf

numeric specifying the confidence level of the simulated envelopes. Default is conf = 0.95.

n.sim

a positive integer representing the number of iterations to generate the simulated envelopes. Default is n.sim = 100.

PrgBar

logical. If PrgBar = TRUE the progress bar will be shown in the console. Default is PrgBar = TRUE.

control

a list of control arguments specified via robustbetareg.control.

...

arguments passed to plot.

Details

The plotenvelope creates normal probability plots with simulated envelope (see Atkinson (1985) for details). Under the correct model, approximately 100*conf of the residuals are expected to be inside the envelope.

Value

plotenvelope returns normal probability plot of residuals with simulated envelope.

Author(s)

Yuri S. Maluf ([e-mail]), Francisco F. Queiroz ([e-mail]) and Silvia L. P. Ferrari.

References

Maluf, Y.S., Ferrari, S.L.P., and Queiroz, F.F. (2022). Robust beta regression through the logit transformation. arXiv:2209.11315.

Atkinson, A.C. (1985) Plots, transformations and regression: an introduction to graphical methods of diagnostic regression analysis. Oxford Science Publications, Oxford.

See Also

robustbetareg, robustbetareg.control, residuals.robustbetareg

Examples

get(data("HIC", package = "robustbetareg"))
hic <- robustbetareg(HIC ~ URB + GDP | GDP,
data = HIC, alpha = 0.06)
plotenvelope(hic, n.sim = 50)

get(data("Firm", package = "robustbetareg"))
rmc <- robustbetareg(FIRMCOST ~ INDCOST + SIZELOG | INDCOST + SIZELOG, data = Firm)
plotenvelope(rmc, conf = 0.90)