qqresid: QQ-plots of DPIT residuals

In assessor: Assessment Tools for Regression Models with Discrete and Semicontinuous Outcomes

QQ-plots of DPIT residuals

Description

Makes a QQ-plot of the DPIT residuals calculated from resid_disc(), resid_semiconti() or resid_zeroinfl(). The plot should be close to the diagonal if the model is correctly specified. Note that this function does not return residuals. To get both residuals and QQ-plot, use resid_disc(), resid_semiconti() and resid_zeroinfl().

Usage

qqresid(model, scale="normal")

Arguments

model	Fitted model object (e.g., glm(), glm.nb(), zeroinfl(), and polr())
scale	You can choose the scale of the residuals between normal and uniform scales. The sample quantiles of the residuals are plotted against the theoretical quantiles of a standard normal distribution under the normal scale, and against the theoretical quantiles of a uniform (0,1) distribution under the uniform scale. The defalut scale is normal.

Value

A QQ plot.

• x-axis: Theoretical quantiles

• y-axis: Sample quantiles generated by DPIT residuals

See Also

resid_disc(), resid_semiconti(), resid_zeroinfl()

Examples

n <- 100
b <- c(2, 1, -2)
x1 <- rnorm(n)
x2 <- rbinom(n, 1, 0.7)
y <- rpois(n, exp(b[1] + b[2] * x1 + b[3] * x2))

m1 <- glm(y ~ x1 + x2, family = poisson)
qqresid(m1, scale = "normal")
qqresid(m1, scale = "uniform")

assessor documentation built on April 3, 2025, 9:21 p.m.