plotvglm.RdCurrently this function plots the Pearson residuals versus the linear predictors (\(M\) plots) and plots the Pearson residuals versus the hat values (\(M\) plots).
plotvglm(x, which = "(All)", ...)An object of class "vglm" (see vglm-class)
or inherits from that class.
If a subset of the plots is required, specify a subset of the
numbers 1:(2*M).
The default is to plot them all.
Arguments fed into the primitive plot
functions.
This function is under development.
Currently it plots the Pearson residuals
against the predicted
values (on the transformed scale) and the hat values.
There are \(2M\) plots in total, therefore
users should call par
to assign, e.g., the mfrow argument.
Note: Section 3.7 of Yee (2015) describes the
Pearson residuals and hat values for VGLMs.
Returns the object invisibly.
if (FALSE) { # \dontrun{
ndata <- data.frame(x2 = runif(nn <- 200))
ndata <- transform(ndata, y1 = rnbinom(nn, mu = exp(3+x2), size = exp(1)))
fit1 <- vglm(y1 ~ x2, negbinomial, data = ndata, trace = TRUE)
coef(fit1, matrix = TRUE)
par(mfrow = c(2, 2))
plot(fit1)
# Manually produce the four plots
plot(fit1, which = 1, col = "blue", las = 1, main = "main1")
abline(h = 0, lty = "dashed", col = "gray50")
plot(fit1, which = 2, col = "blue", las = 1, main = "main2")
abline(h = 0, lty = "dashed", col = "gray50")
plot(fit1, which = 3, col = "blue", las = 1, main = "main3")
plot(fit1, which = 4, col = "blue", las = 1, main = "main4")
} # }