pool.jack: Jackknife Standard Errors and Confidence Intervals

In jmedzihorsky/pistar: Rudas, Clogg and Lindsay Mixture Index of Fit

Jackknife Standard Errors and Confidence Intervals

Description

pool.jack is used to obtain standard errors and confidence intervals from the output of jackknife procedures.

Usage

pool.jack(data, ct = TRUE, estimate, jack_est, side = NULL, conf = 0.95, 
          lower = -Inf, upper = Inf, bias = FALSE)

Arguments

data	the data.
ct	logical: is the data a contingency table?
estimate	the value of the estimate.
jack_est	the vector of estimates from jackknife replications.
side	NULL for a two-sided c.i., "lower" or "upper" for a one-sided interval. "auto" selects "lower" if the estimate is larger than 0 and "upper" if it is smaller than 0.
conf	confidence level expressed as a number between 0 and 1. e.g. for a 95% confidence interval supply 0.95
lower	lowest possible value for one-sided c.i.
upper	largest possible value for one-sided c.i.
bias	logical: apply bias correction? Bias correction currently not implemented.

Value

A named list with the following components:

se	standard error.
low	lower endpoint of c.i.
upp	upper endpoint of c.i.
conf	confidence level of the c.i.
side	"lower" and "upper" for one-sided c.i., "both" for a two-sided c.i.
bias	implementation of bias correction not finished.

Author(s)

Juraj Medzihorsky

References

Efron, Bradley, and Robert Tibshirani. (1993). An introduction to the bootstrap. Vol. 57. CRC press.

jmedzihorsky/pistar documentation built on June 4, 2022, 9:58 a.m.