boot.pwexpm: Bootstrap a Piecewise Exponential Model

Description

Bootstrap an existing piecewise exponential model or build a piecewise exponential model with bootstrapping.

Usage

# S3 method for default
boot.pwexpm(Surv, data, nsim=100, breakpoint=NULL, nbreak=0,
               exclude_int=NULL, min_pt_tail=5, max_set=1000, seed=1818,
               optimizer='mle', tol=1e-4, parallel=FALSE, mc.core=4, ...)
# S3 method for pwexpm
boot.pwexpm(Surv, nsim=100, max_set=1000, seed=1818,
               optimizer='mle', tol=1e-4, parallel=FALSE, mc.core=4, ...)

Value

A object of class "boot.pwexpm" is a list containing the following components:

brk: estimated breakpoints in each row.
lam: estimated piecewise hazard rates in each row.
logLik: the log-likelihood of the original model.
AIC: the Akaike information criterion of the original model.
BIC: the Bayesian information criterion of the original model.
para: the parameters used to estimate the model.

The plot function can be used to make a simple plot for boot.pwexpm.

Arguments

Surv: a Surv object indicating event time and status or a pwexpm object.
data: a data frame in which to interpret the variables named in the Surv argument.
nsim: the number of repeated bootstrapping.
breakpoint: pre-specified breakpoints. See pwexpm.
nbreak: total number of breakpoints. See pwexpm.
exclude_int: an interval that excludes any estimated breakpoints. See pwexpm.
min_pt_tail: the minimum number of events used for estimating the tail (the hazard rate of the last piece). See pwexpm.
max_set: maximum estimated combination of breakpoints. See pwexpm.
seed: a random seed. Do not set seed if seed=NULL.
optimizer: one of the optimizers: mle, ols, or hybrid. See pwexpm.
tol: the minimum allowed gap between two breakpoints. The gap is calculated as (max(time)-min(time))*tol. Keep it as default in most cases.
parallel: logical. If TRUE, use doSNOW package to run in parallel.
mc.core: number of processes allowed to be run in parallel.
...: internal function reserved.

Author

Tianchen Xu zjph602xutianchen@gmail.com

Details

Use bootstrap to repeatdly call pwexpm to estimate the uncertainty of parameters.

Examples

Run this code

event_dist <- function(n)rpwexpm(n, rate = c(0.1, 0.01, 0.2), breakpoint =  c(5,14))
dat <- simdata(rand_rate = 20, drop_rate = 0.03,  total_sample = 1000,
               advanced_dist = list(event_dist=event_dist),
               add_column = c('censor_reason','event','followT','followT_abs'))

fit_res3 <- pwexpm(Surv(followT, event), data = dat, nbreak = 2)
fit_res_boot <- boot.pwexpm(fit_res3, nsim = 10) # here nsim=10 is for demo purpose,
                                                  # pls increase it in practice
plot_survival(dat$followT, dat$event, xlim=c(0,40))
plot_survival(fit_res_boot, col='red', CI_par = list(col='red'))
brk_ci <- apply(fit_res_boot$brk, 2, function(x)quantile(x,c(0.025,0.975)))
abline(v=brk_ci, col='grey', lwd=2)

Run the code above in your browser using DataLab