residuals {ARCensReg} | R Documentation |
Extract model residuals from ARpCRM or ARtpCRM objects
Description
The conditional residuals are obtained by subtracting the fitted values from the response vector, while the quantile residuals are obtained by inverting the estimated distribution function for each observation to obtain approximately normally distributed residuals. See, for instance, Dunn and Smyth (1996) and Kalliovirta (2012).
Usage
## S3 method for class 'ARpCRM'
residuals(object, ...)
## S3 method for class 'ARtpCRM'
residuals(object, ...)
Arguments
object |
An object inheriting from class |
... |
Further arguments passed to or from other methods. |
Value
An object of class "residARpCRM", with the following components:
residuals |
Vector with the conditional residuals of length |
quantile.resid |
Vector with the quantile residuals of length |
Generic function plot
has methods to show a graphic of residual vs. time, an autocorrelation plot, a histogram, and Quantile-Quantile (Q-Q) plot for the quantile residuals.
Author(s)
Fernanda L. Schumacher, Katherine L. Valeriano, Victor H. Lachos, Christian E. Galarza, and Larissa A. Matos
References
Dunn PK, Smyth GK (1996). “Randomized quantile residuals.” Journal of Computational and Graphical Statistics, 5(3), 236–244.
Kalliovirta L (2012). “Misspecification tests based on quantile residuals.” The Econometrics Journal, 15(2), 358–393.
See Also
Examples
## Example 1: Generating data with normal innovations
set.seed(93899)
x = cbind(1, runif(300))
dat1 = rARCens(n=300, beta=c(1,-1), phi=c(.48,-.2), sig2=.5, x=x,
cens='left', pcens=.05, innov="norm")
# Fitting the model with normal innovations
mod1 = ARCensReg(dat1$data$cc, dat1$data$lcl, dat1$data$ucl, dat1$data$y,
x, p=2, tol=0.001)
mod1$tab
plot(residuals(mod1))
# Fitting the model with Student-t innovations
mod2 = ARtCensReg(dat1$data$cc, dat1$data$lcl, dat1$data$ucl, dat1$data$y,
x, p=2, tol=0.001)
mod2$tab
plot(residuals(mod2))
## Example 2: Generating heavy-tailed data
set.seed(12341)
x = cbind(1, runif(300))
dat2 = rARCens(n=300, beta=c(1,-1), phi=c(.48,-.2), sig2=.5, x=x,
cens='left', pcens=.05, innov="t", nu=3)
# Fitting the model with normal innovations
mod3 = ARCensReg(dat2$data$cc, dat2$data$lcl, dat2$data$ucl, dat2$data$y,
x, p=2, tol=0.001)
mod3$tab
plot(residuals(mod3))
# Fitting the model with Student-t innovations
mod4 = ARtCensReg(dat2$data$cc, dat2$data$lcl, dat2$data$ucl, dat2$data$y,
x, p=2, tol=0.001)
mod4$tab
plot(residuals(mod4))