broom (version 0.4.0)

boot_tidiers: Tidying methods for bootstrap computations

Description

Tidying methods for "boot" objects from the "boot" package.

Usage

## S3 method for class 'boot':
tidy(x, conf.int = FALSE, conf.level = 0.95,
  conf.method = "perc", ...)

Arguments

x

boot object

conf.int

whether to include a confidence interval

conf.level

confidence level for CI

conf.method

method for computing confidence intervals (see boot.ci)

...

extra arguments (not used)

Value

• The tidy method returns a data frame with one row per bootstrapped statistic that was calculated, and the following columns:
• term
  Name of the computed statistic, if present
• statistic
  The original values of the statistic
• bias
  The bias of the original statistic value
• std.error
  Standard error of the statistic
• If weights were provided to the boot function, an estimate column is included showing the weighted bootstrap estimate, and the standard error is of that estimate.

  If there are no original statistics in the "boot" object, such as with a call to tsboot with orig.t = FALSE, the original and statistic columns are omitted, and only estimate and std.error columns shown.

Examples

if (require("boot")) {
   clotting <- data.frame(
          u = c(5,10,15,20,30,40,60,80,100),
          lot1 = c(118,58,42,35,27,25,21,19,18),
          lot2 = c(69,35,26,21,18,16,13,12,12))

   g1 <- glm(lot2 ~ log(u), data = clotting, family = Gamma)

   bootfun <- function(d, i) {
      coef(update(g1, data= d[i,]))
   }
   bootres <- boot(clotting, bootfun, R = 999)
   tidy(g1, conf.int=TRUE)
   tidy(bootres, conf.int=TRUE)
}