Use bootstrap output to get CI

Share:

Description

Given bootstrap output of functions such as lllcrc or vgam.crc, produce confidence intervals corresponding to a selection of covariate vectors.

Usage

1
2
extract.CI(mod, probs = c(0.025, 0.975), cont.var = "x.con.1",
  selection = NULL)

Arguments

mod

The object that is output by a lllcrc or vgam.crc.

probs

A vector c(a,b), with 0<a<b<1, marking the upper and lower probability bounds of the desired confidence interval

cont.var

The name of one continuous covariate to serve as the primary index of the ordered set of confidence intervals (one CI for each covariate vector)

selection

A set of categorical covariate names. CIs will be produced only for covariate vectors that have a one in every position corresponding to the elements of selection.

Value

A data frame that includes cont.var, the corresponding point estimate of the rate of missingness pi0, and the lower and upper bounds of the associated confidence intervals.

Author(s)

Zach Kurtz

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.