The exactMatch
function creates a smaller matching problem by
stratifying observations into smaller groups. For a problem that is larger
than maximum allowed size, minExactMatch
provides a way to find the
smallest exact matching problem that will allow for matching.
minExactMatch(x, scores = NULL, width = NULL, maxarcs = 1e+07, ...)
A factor grouping units, suitable for exactMatch
.
The object for dispatching.
Optional vector of scores that will be checked against a caliper width.
Optional width of a caliper to place on the scores.
The maximum problem size to attempt to fit.
Additional arguments for methods.
x
is a formula of the form Z ~ X1 + X2
, where
Z
is indicates treatment or control status, and X1
and X2
are variables
can be converted to factors. Any additional arguments are passed to model.frame
(e.g., a data
argument containing Z
, X1
, and X2
).
The the arguments scores
and width
must be passed together.
The function will apply the caliper implied by the scores and the width while
also adding in blocking factors.