Skip to contents

Write out the equation of a linear model

Source: R/lm_equation.R
lm_equation.Rd

Takes a fitted linear model and returns a human-readable equation string showing the relationship between the response variable and predictors. Coefficients are rounded to 3 significant figures for readability. Terms are named as in the fitted coefficients, so factor predictors show one term per dummy level (e.g. 2.1*gB) and models without an intercept print no intercept.

Usage

lm_equation(model, style = c("prettier", "brackets", "raw"))

Arguments

model

A linear model

style

How factor terms are named. "prettier" (the default) spells out the factor name and level (e.g. 4.09*(Species="setosa")), "brackets" shows the level alone (e.g. 4.09*[setosa]), and "raw" keeps the design-matrix names (e.g. 4.09*Speciessetosa).

Value

A character string of length 1 containing the fitted equation, e.g. "mpg = 30.7 - 0.0248*disp - 0.0245*hp".

Details

Use it when you want to report or sanity-check a fitted model as an equation rather than a coefficient table — for example when writing up homework or checking which dummy terms a factor produced.

Examples

# Simple regression
lm_equation(lm(mpg ~ wt, data = mtcars))
#> [1] "mpg = 37.3 - 5.34*wt"

# Multiple predictors
lm_equation(lm(mpg ~ disp + hp, data = mtcars))
#> [1] "mpg = 30.7 - 0.0303*disp - 0.0248*hp"

# Factor predictor: one term per non-reference level, spelled out by default
model <- lm(Sepal.Length ~ Sepal.Width + Species, data = iris)
lm_equation(model)
#> [1] "Sepal.Length = 2.25 + 0.804*Sepal.Width + 1.46*(Species=\"versicolor\") + 1.95*(Species=\"virginica\")"

# `style = "brackets"` shows just the level; `"raw"` keeps design-matrix names
lm_equation(model, style = "brackets")
#> [1] "Sepal.Length = 2.25 + 0.804*Sepal.Width + 1.46*[versicolor] + 1.95*[virginica]"
lm_equation(model, style = "raw")
#> [1] "Sepal.Length = 2.25 + 0.804*Sepal.Width + 1.46*Speciesversicolor + 1.95*Speciesvirginica"

# Transformed terms and interactions
lm_equation(lm(mpg ~ wt + I(wt^2), data = mtcars))
#> [1] "mpg = 49.9 - 13.4*wt + 1.17*I(wt^2)"
lm_equation(lm(Sepal.Length ~ Sepal.Width * Species, data = iris))
#> [1] "Sepal.Length = 2.64 + 0.69*Sepal.Width + 0.901*(Species=\"versicolor\") + 1.27*(Species=\"virginica\") + 0.175*Sepal.Width:(Species=\"versicolor\") + 0.211*Sepal.Width:(Species=\"virginica\")"

# No-intercept models print no intercept
lm_equation(lm(mpg ~ 0 + wt, data = mtcars))
#> [1] "mpg = 5.29*wt"