fc_boot {failCompare}R Documentation

Nonparametric bootstrap of failure time model object

Description

Nonparametric bootstrap of failure time model object

Usage

fc_boot(mod_obj, nrep, type = "pred", times = NULL, tol = 0.9, ...)

Arguments

mod_obj

failure time model object of class "fc_obj"

nrep

number of resampling replicates

type

character describing whether bootstrap samples of predicted survivals ("pred") or parameters ("par") should be returned.

times

times at which survival fraction will be estimated, if type="pred".

tol

optional tolerance setting for the estimated proportion of bootstrap data sets that cannot be fit, default = 0.9

...

arguments passed to the optimizer

Details

Random sampling of the failure time data with replacement as a means for propagating uncertainty in predictions of survival probability and estimates of parameter sampling distributions (Tibshirani and Efron 1993).

Value

if type="pred" survival fraction or proportion of failed subjects (nrep x times) is returned, and if type="par" a matrix of bootstrap parameter estimates dimensions (nrep x number of parameters).

References

Efron, B. and Tibshirani, R. 1993 An Introduction to the Bootstrap. Chapman and Hall, New York.

Townsend R., L., J. R. Skalski, P. Dillingham, T. W. Steig. 2006 Correcting bias in survival estimation resulting from tag failure in acoustic and radiotelemetry studies. Journal of Agricultural Biological and Environmental Statistics.11:183-196.

See Also

fc_fit

Examples

data(sockeye)
taglife=sockeye[,"days"]
weib_mod=fc_fit(taglife,model="weibull")
fc_boot(weib_mod,nrep=60,times = 10:20)


[Package failCompare version 1.0.0 Index]