Additional Arguments for Nearest Neighbor Matching

• m.order: the order in which to match treatment units with control units.
  • "largest" (default): matches from the largest value of the distance measure to the smallest.
  • "smallest": matches from the smallest value of the distance measure to the largest.
  • "random": matches in random order.
• replace: logical value indicating whether each control unit can be matched to more than one treated unit (default = replace = FALSE, each control unit is used at most once - i.e., sampling without replacement). For matching with replacement, use replace = TRUE. After matching with replacement, the weights can be used to reflect the frequency with which each control unit was matched.
• ratio: the number of control units to match to each treated unit (default = 1). If matching is done without replacement and there are fewer control units than ratio times the number of eligible treated units (i.e., there are not enough control units for the specified method), then the higher ratios will have NA in place of the matching unit number in match.matrix.
• exact: variables on which to perform exact matching within the nearest neighbor matching (default = NULL, no exact matching). If exact is specified, only matches that exactly match on the covariates in exact will be allowed. Within the matches that match on the variables in exact, the match with the closest distance measure will be chosen. exact should be entered as a vector of variable names (e.g., exact = c("X1", "X2")).
• caliper: the number of standard deviations of the distance measure within which to draw control units (default = 0, no caliper matching). If a caliper is specified, a control unit within the caliper for a treated unit is randomly selected as the match for that treated unit. If caliper != 0, there are two additional options:
  • calclosest: whether to take the nearest available match if no matches are available within the caliper (default = FALSE).
  • mahvars: variables on which to perform Mahalanobis-metric matching within each caliper (default = NULL). Variables should be entered as a vector of variable names (e.g., mahvars = c("X1", "X2")). If mahvars is specified without caliper, the caliper is set to 0.25.
• subclass and sub.by: See the options for subclassification for more details on these options. If a subclass is specified within method = "nearest", the matched units will be placed into subclasses after the nearest neighbor matching is completed.