predab.resample is a general-purpose
function that is used by functions for specific models.
It computes estimates of optimism of, and bias-corrected estimates of a vector
of indexes of predictive accuracy, for a model with a specified
design matrix, with or without fast backward step-down of predictors. If
bw=TRUE, the design
x must have been created by
predab.resample stores as the
attribute a logical matrix encoding which
factors were selected at each repetition.
predab.resample(fit.orig, fit, measure, method=c("boot","crossvalidation",".632","randomization"), bw=FALSE, B=50, pr=FALSE, prmodsel=TRUE, rule="aic", type="residual", sls=.05, aics=0, tol=1e-12, force=NULL, estimates=TRUE, non.slopes.in.x=TRUE, kint=1, cluster, subset, group=NULL, allow.varying.intercepts=FALSE, debug=FALSE, …)
object containing the original full-sample fit, with the
y=TRUE options specified to the model fitting function. This model
should be the FULL model including all candidate variables ever excluded
because of poor associations with the response.
a function to fit the model, either the original model fit, or a fit in a
sample. fit has as arguments
xcol, and other arguments passed to
If you don't want
as an argument inside the definition of
fit, add … to the end of its
iter is passed to
fit to inform the function of the
sampling repetition number (0=original sample). If
allow for the possibility of selecting no predictors, i.e., it should fit an
intercept-only model if the model has intercept(s).
fit must return
fit failed due to singularity or
non-convergence - these cases are excluded from summary statistics).
must add design attributes to the returned object if
penalty.matrix parameter is not used if
vector is a vector of columns of
X to be used in the current model fit.
psm it includes a
1 for the intercept position.
xcol is not defined if
iter=0 unless the initial fit had been from
a backward step-down.
xcol is used to select the correct rows and columns
penalty.matrix for the current variables selected, for example.
a function to compute a vector of indexes of predictive accuracy for a given fit.
method="crossval", it will make the most sense for
measure to compute only indexes that are independent of sample size. The
measure function should take the following arguments or use …:
(X beta for
iter is as in
evalfit is set to
predab.resample if the fit is being evaluated on the sample used to make the
fit.orig is the fit object returned by the original fit on the whole
evalfit will sometimes save computations. For example, in
bootstrapping the area under an ROC curve for a logistic regression model,
lrm already computes the area if the fit is on the training sample.
is used to pass computed configuration parameters from the original fit such as
quantiles of predicted probabilities that are used as cut points in other samples.
The vector created by measure should have
names() associated with it.
The default is
"boot" for ordinary bootstrapping (Efron, 1983,
Eq. 2.10). Use
".632" for Efron's
.632 method (Efron,
1983, Section 6 and Eq. 6.10),
"crossvalidation" for grouped
"randomization" for the randomization
method. May be abbreviated down to any level, e.g.
TRUE to do fast backward step-down for each training
sample. Default is
Number of repetitions, default=50. For
this is also the number of groups the original sample is split into.
TRUE to print results for each sample. Default is
FALSE to suppress printing of model selection output such
as that from
Stopping rule for fastbw,
"p". Default is
"aic" to use Akaike's information criterion.
Type of statistic to use in stopping rule for fastbw,
(the default) or
Significance level for stopping in fastbw if
rule="p". Default is
Stopping criteria for
rule="aic". Stops deleting factors when
chi-square - 2 times d.f. falls below
aics. Default is
Tolerance for singularity checking. Is passed to
FALSE if the design matrix
does not have columns for intercepts and these columns are needed
For multiple intercept models such as the ordinal logistic model, you may
specify which intercept to use as
kint. This affects the linear
predictor that is passed to
Vector containing cluster identifiers. This can be specified only if
method="boot". If it is present, the bootstrap is done using sampling
with replacement from the clusters rather than from the original records.
If this vector is not the same length as the number of rows in the data
matrix used in the fit, an attempt will be made to use
fit.orig to conform
cluster to the data.
bootcov for more about this.
specify a vector of positive or negative integers or a logical vector when
you want to have the
measure function compute measures of accuracy on
a subset of the data. The whole dataset is still used for all model development.
For example, you may want to
calibrate a model by
assessing the predictions on females when the fit was based on males and
females. When you use
cr.setup to build extra observations for fitting the
continuation ratio ordinal logistic model, you can use
subset to specify
cohort or observations to use for deriving indexes of predictive
accuracy. For example, specify
subset=cohort=="all" to validate the
model for the first layer of the continuation ratio model (Prob(Y=0)).
a grouping variable used to stratify the sample upon bootstrapping. This allows one to handle k-sample problems, i.e., each bootstrap sample will be forced to selected the same number of observations from each level of group as the number appearing in the original dataset.
TRUE to not throw an error
if the number of intercepts varies from fit to fit
TRUE to print subscripts of all training and
The user may add other arguments here that are passed to
a matrix of class
"validate" with rows corresponding
to indexes computed by
measure, and the following columns:
indexes in original overall fit
average indexes in training samples
average indexes in test samples
training-test except for
method=".632" - is .632 times
(index.orig - test)
number of successful repetitions with the given index non-missing
method=".632", the program stops with an error if every observation
is not omitted at least once from a bootstrap sample. Efron's ".632" method
was developed for measures that are formulated in terms on per-observation
contributions. In general, error measures (e.g., ROC areas) cannot be
written in this way, so this function uses a heuristic extension to
Efron's formulation in which it is assumed that the average error measure
ith observation is the same as the average error measure
omitting any other observation. Then weights are derived
for each bootstrap repetition and weighted averages over the
can easily be computed.
Efron B, Tibshirani R (1997). Improvements on cross-validation: The .632+ bootstrap method. JASA 92:548--560.