compare_models.lm: Compare two fitted linear models

In dScoreTest: Debiased Score Tests for Goodness of Fit and Model Comparison

Compare two fitted linear models

Description

Debiased score test of the null fit.0 against the alternative fit.1. Both are internally refit as Gaussian-family GLMs and the call is dispatched to compare_models.glm.

Usage

## S3 method for class 'lm'
compare_models(fit.0, fit.1, ...)

Arguments

fit.0	The null model as a fitted lm object.
fit.1	The alternative model as a fitted lm object. Must be a supermodel of fit.0; see compare_models.glm.
...	Additional arguments passed to compare_models.glm.

Value

An object of class "dScoreTest": a list whose key elements are the debiased test statistic t.stat and the one-sided p-value p.val (right tail of the standard normal), along with the test-set score residuals, the hunted direction, and the call. It has print, summary and plot methods.

Examples

 set.seed(42)
 n <- 500
 dat <- data.frame(x1 = rnorm(n), x2 = rnorm(n), x3 = rnorm(n))
 dat$x3 <- dat$x3 + (dat$x1 + dat$x2) / 3
 dat$y  <- 1 + dat$x1 + 2 * dat$x3 + rnorm(n)
 fit.0 <- lm(y ~ x1 + x3,        data = dat)
 fit.1 <- lm(y ~ x1 + x2 + x3,   data = dat)
 
 # test fit.0 against fit.1: should not be rejected
 compare_models(fit.0, fit.1)
 anova(fit.0, fit.1)

 # misspecified model: should be rejected
 fit.00 <- lm(y ~ x1 + x2, data = dat)
 compare_models(fit.00, fit.1)
 plot(compare_models(fit.00, fit.1))
 compare_models(fit.00, fit.1, hunt.style="wls")
 anova(fit.00, fit.1)

dScoreTest documentation built on Sept. 3, 2026, 1:06 a.m.