Learn R Programming

optmatch (version 0.9-10)

maxControlsCap: Set thinning and thickening caps for full matching

Description

Functions to find the largest value of min.controls, or the smallest value of max.controls, for which a full matching problem is feasible. These are determined by constraints embedded in the matching problem's distance matrix.

Usage

maxControlsCap(distance, min.controls = NULL)

minControlsCap(distance, max.controls = NULL)

Arguments

distance

Either a matrix of non-negative, numeric discrepancies, or a list of such matrices. (See fullmatch for details.)

min.controls

Optionally, set limits on the minimum number of controls per matched set. (Only makes sense for maxControlsCap.)

max.controls

Optionally, set limits on the maximum number of controls per matched set. (Only makes sense for minControlsCap.)

Value

For minControlsCap, strictest.feasible.min.controls and given.max.controls. For maxControlsCap, given.min.controls and strictest.feasible.max.controls.

strictest.feasible.min.controls

The largest values of the fullmatch argument min.controls that yield a full match;

given.max.controls

The max.controls argument given to minControlsCap or, if none was given, a vector of Infs.

given.min.controls

The min.controls argument given to maxControlsCap or, if none was given, a vector of 0s;

strictest.feasible.max.controls

The smallest values of the fullmatch argument max.controls that yield a full match.

Details

The function works by repeated application of full matching, so on large problems it can be time-consuming.

References

Hansen, B.B. and S. Olsen Klopfer (2006), ‘Optimal full matching and related designs via network flows’, Journal of Computational and Graphical Statistics 15, 609--627.

See Also

fullmatch