Bootstrapping for regression models
This function provides a simple front-end to the
boot function in the
package also called
boot is very general and therefore
has many arguments, the
Boot function has very few arguments, but should
meet the needs of many users.
## S3 method for class 'default': Boot(object, f=coef, labels=names(coef(object)), R=999, method=c("case", "residual")) ## S3 method for class 'lm': Boot(object, f=coef, labels=names(coef(object)), R=999, method=c("case", "residual")) ## S3 method for class 'glm': Boot(object, f=coef, labels=names(coef(object)), R=999, method=c("case", "residual"))
- A regression object of class
glm. The function may work with other regression objects that support the
updatemethod and have a
subsetargument, but it will fail if the fitting method for the mod
- A function whose one argument is the name of a regression object that will be applied to the updated regression object to
compute the statistics of interest. The default is
coef, to return to regression coefficient estimates. For example,
- Provides labels for the statistics computed by
f. If this argument is of the wrong length, then generic labels will be generated.
- Number of bootstrap samples.
- The bootstrap method, either
casefor resampling cases or residualsfor a residual bootstrap. See the details below. The residual bootstrap is available only for
lmobjects and will return an error for
boot function is
Boot is very specific. It takes the information from a
regression object and the choice of
method, and creates a function that is
passed as the
statistic argument to
boot. The argument
is also passed to
boot. All other arguments to
kept at their default values.
The methods available for
lm objects are
This function may fail if the model fit to any of the bootstrap samples is of lower rank than the model fit to the original data. This will occur, for example, if the model includes factors and interactions with a very small number of observations per cell. In this case bootstrap samples may have zero counts in some cells and lose rank.
bootfor the returned value from this function. The car package includes additional generic functions, as listed below.
Davison, A, and Hinkley, D. (1997) Bootstrap Methods and their
Applications. Oxford: Oxford University Press.
Fox, J. and Weisberg, S. (2011) Companion to Applied Regression, Second Edition.
Thousand Oaks: Sage.
Fox, J. and Weisberg, S. (2012) Bootstrapping,
m1 <- lm(Fertility ~ ., swiss) betahat.boot <- Boot(m1, R=99) # 99 bootstrap samples--too small to be useful summary(betahat.boot) # default summary confint(betahat.boot) hist(betahat.boot) # Bootstrap for the estimated residual standard deviation: sigmahat.boot <- Boot(m1, R=99, f=sigmaHat, labels="sigmaHat") summary(sigmahat.boot) confint(sigmahat.boot)