get_jackknife_values: Jackknife values
In disprofas: Non-Parametric Dissolution Profile Analysis

Description Usage Arguments Details Value References See Also

The function get_jackknife_values() is required for generation of vector L for the BCa confidence interval estimation by the boot.ci() function from the ‘boot’ package.

1	get_jackknife_values(grouping, stat_fun, data, ...)

`grouping`	A character string specifying the column in `data` that contains the group names (i.e. a factorial variable, e.g., for the differentiation of batches or formulations of a drug product).
`stat_fun`	The name of the function calculating the statistic of interest. The statistic function must return a single number.
`data`	A data frame with the dissolution profile data in wide format and a column for the distinction of the groups to be compared.
`...`	Further arguments that may be passed down to the statistics function.

For the calculation of BCa bootstrap confidence intervals empirical influence values of the statistic of interest for the observed data are required. If L is not provided to the boot.ci() function from the ‘boot’ package, the values are calculated, if needed, using the empinf() function (also from the ‘boot’ package). Since the results returned by empinf() are not appropriate in this case, the L values are determined using get_jackknife_values(). The jackknife calculations are done by calculating the leave-one-out estimates.

A list of the jackknife assessment with the following elements is returned:

`theta.hat`	The original value of the statistic.
`theta.jack`	The jackknife value of the statistic.
`jack.se`	The jackknife standard error of the statistic.
`jack.bias`	The bias of the statistic.
`loo.values`	The leave-one-out values.
`pseudo.values`	The pseudo values.