lm: Fit a linear model

In cobriant/qelp: Quick Help: Beginner friendly help docs for R and the Tidyverse

Fit a linear model

Description

Run a regression using ordinary least squares.

Usage

lm(formula, data)

Arguments

formula	A symbolic description of the model. For example: y ~ x estimates the model y = beta_0 + beta_1*x + u.
data	A data frame like a tibble.

Examples

# A tibble with the heights and weights
# of 5 individuals:

library(tidyverse)

htwt <- tibble(
  sex = c("male", "male", "female", "female", "male"),
  height = c(62, 72, 68, 64, 70),
  weight = c(169, 243, 147, 135, 183)
)

# Simple regression:

lm(weight ~ height, data = htwt)

# Simple regression with tidied output:

lm(weight ~ height, data = htwt) %>%
  broom::tidy(conf.int = TRUE)

-----------------------------------

# Multiple regression:

lm(weight ~ height + sex, data = htwt)

# Multiple regression without an intercept:

lm(weight ~ 0 + height + sex, data = htwt)

-----------------------------------

# Interaction terms: ':' versus '*'
#   Use ':' to include only an interaction.
#   Use '*' to also include the terms alone.
#   So 'x*z' is equivalent to 'x + z + x:z'
#   And `x*y*z` is equivalent to
#     'x + y + z + x:y + x:z + y:z + x:y:z'

lm(weight ~ height:sex, data = htwt)

lm(weight ~ height*sex, data = htwt)

-----------------------------------

# Transformations of variables:
#  To include a term squared, use I()
#  to first parse the expression
#  before fitting the model.

lm(weight ~ I(height^2), data = htwt)

-----------------------------------

# Fitted values and residuals

lm(weight ~ height, data = htwt) %>%
  fitted.values()

lm(weight ~ height, data = htwt) %>%
  residuals()

-----------------------------------

# R squared

lm(weight ~ height, data = htwt) %>%
  broom::glance() %>%
  select(r.squared)

cobriant/qelp documentation built on July 1, 2022, 7:24 a.m.