Compute the test statistic for election audits, essentially the largest error found in the audit, as measured by the passed functions and methods.
compute.stark.t(
Z,
bound.col,
calc.e_p = calc.pairwise.e_p,
w_p = weight.function("no.weight"),
err.override = NULL,
return.revised.audit = FALSE
)If it already has an audit table with err and err.weighted then it will use those errors, otherwise it will compute them with compute.stark.err
This is the vector containing the maximum number of votes possible in the various precincts.
Function to compute e_p. Default is calc.pairwise.e_p.
The weight function to be applied to the precinct error.
If non-null, use this as the found error in votes rather than the actual errors found in the audit.
Return the updated audit frame with the error and weighted errors calculated.
The test statistic, i.e. the maximum found error in the audit sample, as computed by calc.e\_p and weighted by w\_p.
This is an older method that other methods sometime use---it is probably best ignored unless you have a good reason not to.