proc psmatch question: Is K= option same as KMAX= option?

[RobertWF1]

For example, if I want to conduct greedy nearest neighbor propensity score matching where 1 to 5 control records may be matched to each treatment record, is specifying K=5 sufficient or will SAS attempt to match exactly 5 controls to each treatment?

The KMAX= option doesn't appear to be available for the greedy matching method.

[sbxkoenk]

I have moved this question to 'Statistical Procedures' board (where it belongs)!

[sbxkoenk · Accepted solution]

---

@RobertWF1 wrote:

For example, if I want to conduct greedy nearest neighbor propensity score matching where 1 to 5 control records may be matched to each treatment record, is specifying K=5 sufficient or will SAS attempt to match exactly 5 controls to each treatment?

The KMAX= option doesn't appear to be available for the greedy matching method.

---

Hello,

I have never done this, but as I read the documentation : 

SAS/STAT® 15.2 User's Guide
The PSMATCH Procedure

https://go.documentation.sas.com/doc/en/statug/15.2/statug_psmatch_syntax08.htm#statug.psmatch.metho...

  METHOD=FULL (KMAX=kmax <full-options>)
  KMAX=kmax >> specifies the maximum number of control units to be matched with each treated unit, where kmax >= 1.

  METHOD=GREEDY <(K=k ORDER=order-option)>

  K=k specifies the number of matching control units, where k > 0, for each treated unit. PROC PSMATCH performs k separate loops of matching for treated units. In each loop, the nearest control unit is sequentially matched to each treated unit. By default, K=1 (one control unit for each treated unit).
AS I READ THIS , IT WILL MATCH EXACTLY K CONTROL UNITS TO EACH TREATED UNIT.

Example 100.4 Greedy Nearest Neighbor Matching
https://go.documentation.sas.com/doc/en/statug/15.2/statug_psmatch_examples04.htm

Best,

Koen