lm.lincom.conf: Confidence Functions for the Expected Response of a Linear...

In ddarmon/clp: R Package for Computing Confidence Functions to Accompany CLP

Description

Confidence functions for the expected response of a linear model under Gaussian noise assumptions.

Usage

lm.lincom.conf(mod, x, plot = TRUE, conf.level = 0.95)

Arguments

mod	an output from lm
x	a vector containing the predictor values for which the expected response is desired
plot	whether to plot the confidence density and curve
conf.level	the confidence level for the confidence interval indicated on the confidence curve

Value

A list containing the confidence functions pconf, dconf, cconf, and qconf for the expected response at x, as well as the P-curve and S-curve.

References

Tore Schweder and Nils Lid Hjort. Confidence, likelihood, probability. Vol. 41. Cambridge University Press, 2016.

Examples

# Prediction of body fat percentage from other body measurements.

data(fat)

mod <- lm(body.fat ~ age + weight + height, data = fat)

# Confidence curve for the expected BFP of a 170lb, 6ft, 20-year old male

x <- c(1,    # Intercept
       20,   # 20 years old
       170,  # 170 lbs
       6*12) # 72 inches (6 feet)

m.conf <- lm.lincom.conf(mod, x)

ddarmon/clp documentation built on Jan. 25, 2021, 6:22 p.m.