tidy.margins: Tidy a(n) margins object

Description

Tidy summarizes information about the components of a model. A model component might be a single term in a regression, a single hypothesis, a cluster, or a class. Exactly what tidy considers to be a model component varies across models but is usually self-evident. If a model has several distinct types of components, you will need to specify which components to return.

Usage

# S3 method for margins
tidy(x, conf.int = FALSE, conf.level = 0.95, ...)

Arguments

A margins object returned from margins::margins().

conf.int

Logical indicating whether or not to include a confidence interval in the tidied output. Defaults to FALSE.

conf.level

The confidence level to use for the confidence interval if conf.int = TRUE. Must be strictly greater than 0 and less than 1. Defaults to 0.95, which corresponds to a 95 percent confidence interval.

...

Additional arguments. Not used. Needed to match generic signature only. Cautionary note: Misspelled arguments will be absorbed in ..., where they will be ignored. If the misspelled argument has a default value, the default value will be used. For example, if you pass conf.level = 0.9, all computation will proceed using conf.level = 0.95. Additionally, if you pass newdata = my_tibble to an augment() method that does not accept a newdata argument, it will use the default value for the data argument.

Value

A tibble::tibble() with columns:

conf.high

Upper bound on the confidence interval for the estimate.

conf.low

Lower bound on the confidence interval for the estimate.

estimate

The estimated value of the regression term.

p.value

The two-sided p-value associated with the observed statistic.

statistic

The value of a T-statistic to use in a hypothesis that the regression term is non-zero.

std.error

The standard error of the regression term.

term

The name of the regression term.

Details

The margins package provides a way to obtain coefficient marginal effects for a variety of (non-linear) models, such as logit or models with multiway interaction terms. Note that the glance.margins() method requires rerunning the underlying model again, which can take some time. Similarly, an augment.margins() method is not currently supported, but users can simply run the underlying model to obtain the same information.

Examples

Run this code

# NOT RUN {
library(margins)

## Example 1: Logit model ##

mod_log <- glm(am ~ cyl + hp + wt, data = mtcars, family = binomial)

# Get tidied "naive" model coefficients
tidy(mod_log)

# Convert to marginal effects with margins::margins()
marg_log <- margins(mod_log)

# Get tidied marginal effects
tidy(marg_log)
tidy(marg_log, conf.int = TRUE)
glance(marg_log) ## Requires running the underlying model again. Quick for this example.
# }
# NOT RUN {
augment(marg_log) ## Not supported.
# }
# NOT RUN {
augment(mod_log) ## But can get the same info by running on the underlying model.

## Example 2: Threeway interaction terms ##

mod_ie <- lm(mpg ~ wt * cyl * disp, data = mtcars)

# Get tidied "naive" model coefficients
tidy(mod_ie)

# Convert to marginal effects with margins::margins()
marg_ie0 <- margins(mod_ie)
# Get tidied marginal effects 
tidy(marg_ie0)
glance(marg_ie0)

# Marginal effects evaluated at specific values of a variable (here: cyl)
marg_ie1 <- margins(mod_ie, at = list(cyl = c(4,6,8)))
tidy(marg_ie1)

# Marginal effects of one interaction variable (here: wt), modulated at 
# specific values of the two other interaction variables (here: cyl and drat)
marg_ie2 <- margins(mod_ie,
                    variables = "wt", ## Main var
                    at = list(cyl = c(4,6,8), drat = c(3, 3.5, 4))) ## Modulating vars
tidy(marg_ie2)
# }