Propensity score matching - code error when using stata help format psmatch2 vs psmatch2 logit

[Martin Imelda Borg]

Question: I am trying to match my propensity scores, but psmatch2 won't work if I use this code format:
(Neighbour hood matching)

(I)
psmatch2 $treatment $xlist, outcome($ylist) common bw(0.06) ps(ps)

Error given:
psmatch2 $treatment $xlist, outcome(infection) common bw(0.06) ps(ps)
You cannot specify both a varlist AND a propensity score
r(198)

(II)
BUT it will work if :
psmatch2 $treatment $xlist, outcome($ylist) n(3) logit


What am I doing wrong? as the first (I) one works on probit and the second one works on logit. I know this depends on my preference (probit vs logit). But I would like to know what I'm doing wrong in the first one (i) why am I getting an error?

Here's a bit of background:

Research question:
Does smoking effect risk of infection after angioplasty?

DATA BELOW:

* Example generated by -dataex-. For more info, type help dataex
clear
input float(smoking infection diabetes age cancer ps u) byte(_n2 _n3)
1 0 0 87 0  .531945 .12884283 1 3
0 0 0 34 0 .7917001   .323584 . .
1 1 0 46 1 .8622774 .49606895 2 1
1 1 0 36 0 .7836868  .6916475 2 1
0 1 1 78 1 .4938836   .901614 . .
1 1 1 45 0 .4986039  .9556306 1 3
1 1 1 44 1 .6752168  .9616644 2 1
0 0 1 76 0 .3329607  .9732198 . .
end

------------

MY CODE:

**Creating macros
global treatment smoking //intervention/treatment
global ylist infection //outcome
global xlist diabetes cancer age //variables to match on t

//Creating propensity scores

probit $treatment $xlist
predict ps

set seed 100 //sorts this out randomly
generate u=runiform()
sort u

//Matching - error given below
psmatch2 $treatment $xlist, outcome($ylist) common bw(0.06) ps(ps)

**This works but is logit not probit

psmatch2 $treatment $xlist, outcome($ylist) n(3) logit

[David Radwin]

Whether you use logit or probit is irrelevant to this problem. See https://www.statalist.org/forums/for...49#post1698349