india (version 0.1-2)

rquantile: Randomized quantile residuals

Description

Compute randomized quantile residuals from a fitted model object.

Usage

rquantile(model, ...)
# S3 method for lm
rquantile(model, ...)
# S3 method for lad
rquantile(model, ...)
# S3 method for ols
rquantile(model, ...)
# S3 method for nls
rquantile(model, ...)
# S3 method for ridge
rquantile(model, ...)

Value

a vector containing standard normal deviates representing standardized residuals. This kind of residuals are exactly normal.

Arguments

model

an R object, returned by lm, lad, ols, nls or ridge.

...

further arguments passed to or from other methods.

References

Dunn, P.K., Smyth, G.K. (1996). Randomized quantile residuals. Journal of Computational and Graphical Statistics 5, 236-244. tools:::Rd_expr_doi("10.1080/10618600.1996.10474708")

Osorio, F. (2026). On the mean-shift outlier model for LAD regression. Working paper.

Examples

# quantile residuals for linear regression
fm <- ols(stack.loss ~ ., data = stackloss)
res <- rquantile(fm)
plot(res, ylim = c(-3,3), ylab = "quantile residuals")
abline(h = 0, lwd = 2, col = "gray75")
abline(h = c(-2,2), lwd = 2, lty = 2, col = "red")
text(21, res[21], as.character(21), pos = 1)

# quantile residuals for LAD regression
data(ereturns)
fm <- lad(m.marietta ~ CRSP, data = ereturns)
res <- rquantile(fm)
plot(res, ylim = c(-2,4.5), ylab = "quantile residuals")
abline(h = 0, lwd = 2, col = "gray75")
abline(h = c(-2,2), lwd = 2, lty = 2, col = "red")
obs <- c(8,15,34)
text(obs, res[obs], as.character(obs), pos = 3)

# quantile residuals for ridge regression
data(portland)
fm <- ridge(y ~ ., data = portland)
res <- rquantile(fm)
plot(res, ylim = c(-2,2), ylab = "quantile residuals")
abline(h = 0, lwd = 2, col = "gray75")

# quantile residuals for nonlinear regression
data(skeena)
model <- recruits ~ b1 * spawners * exp(-b2 * spawners)
fm <- nls(model, data = skeena, start = list(b1 = 3, b2 = 0))
res <- rquantile(fm)
plot(res, ylim = c(-3,3), ylab = "quantile residuals")
abline(h = 0, lwd = 2, col = "gray75")
abline(h = c(-2,2), lwd = 2, lty = 2, col = "red")
text(5, res[5], as.character(5), pos = 3)