add_quantile.lm: Quantiles for the Response of a Linear Model

In ciTools: Confidence or Prediction Intervals, Quantiles, and Probabilities for Statistical Models

Description

This function is one of the methods of add_quantile. It is called automatically when add_quantile is called on objects of class lm.

Usage

## S3 method for class 'lm'
add_quantile(
  df,
  fit,
  p,
  name = NULL,
  yhatName = "pred",
  log_response = FALSE,
  ...
)

Arguments

df	A data frame of new data.
fit	An object of class lm. Predictions are made with this object.
p	A real number between 0 and 1. Sets the level of the quantiles.
name	NULL or a string. If NULL, quantiles automatically will be named by add_quantile, otherwise, they will be named name.
yhatName	A string. Name of the vector of predictions.
log_response	A logical. If TRUE, quantiles will be generated for the prediction made with a log-linear model: \log(Y) = Xβ + ε. These quantiles will be on the scale of the original response, Y.
...	Additional arguments.

Details

Quantiles for linear models are determined parametrically, by applying a pivotal quantity to the distribution of Y|x.

Value

A dataframe, df, with predicted values and level - p quantiles attached.

See Also

add_ci.lm for confidence intervals for lm objects, add_pi.lm for prediction intervals of lm objects, and add_probs.lm for response probabilities of lm objects.

Examples

# Fit a linear Model
fit <- lm(dist ~ speed, data = cars)

# Find the 0.7-quantile (70th percentile) of new distances, given
# the linear model fit.
add_quantile(cars, fit, p = 0.7)

# As above, but with a custom name for the vector of quantiles
add_quantile(cars, fit, p = 0.7, name = "my_quantile")

ciTools documentation built on Jan. 13, 2021, 7 a.m.