ggdiagnose: Graphically diagnose model residuals (ggplot2 version).

In robertschnitman/schnitr: Diagnostic tools for regression modeling

Graphically diagnose model residuals (ggplot2 version).

Description

Graphically diagnose model residuals (ggplot2 version).

Usage

ggdiagnose(model, fit_type = 'response', residual_type = 'response', bins = 30, se = TRUE, freqpct = FALSE, alpha = 1)

Arguments

model	An lm or glm object.
fit_type	String. Default is "response". Type of fitted values to use based on options in predict().
residual_type	String. Default is "response". Type of residuals values to use based on options in resid().
bins	Number of bins to specify for histograms.
se	Boolean. For overlaying shaded standard errors.
freqpct	Boolean.
alpha	Integer, [0, 1]. Points are more transparent the closer they are to 0. Only applies to scatter plots.

Value

2 scatter plots and 2 histograms of residuals and "residuals margin," which is the residuals as a percentage of the actual dependent variable values.

See Also

https://github.com/robertschnitman/diagnoser

Examples

# OLS case
model <- lm(data = mtcars, formula = mpg ~ wt + gear)
ggdiagnose(model, bins = NROW(mtcars), se = FALSE, freqpct = TRUE)

# NLS case
model.nls <- nls(Ozone ~ theta0 + Temp^theta1, airquality)
ggdiagnose(model.nls)

robertschnitman/schnitr documentation built on Aug. 18, 2022, 8:39 p.m.