ignore missing data in proc sgplot histogram

[mateescu]

I have a large set with several variables, some records have missing data for some of the variables. I am trying to generate a histogram (by group) but I get an extra histogram for the missing data (missing data in the group variable). Here is a small example code of what I have. I want to know if there is any way to ignore the missing data in the Group.

Thanks,

data test;
input Var1 Group;
cards;
1 1
2 1
3 1
3 .
4 .
5 .
2 2 
3 2
4 2
5 2
;
run;

title height=16pt "Var by Group";
proc sgplot data=test ; 
   histogram Var1 / group=Group transparency=0.5 datalabelattrs=(size=16) ;
   density Var1 / type=kernel group=Group;
   keylegend /location=outside position=topright valueattrs=(size=14);
   xaxis display=(noline) labelattrs=(size=16) valueattrs=(size=16) ;
  yaxis display=(noline) labelattrs=(size=16) valueattrs=(size=16);

run;
quit;

mateescu_0-1703891292008.png

[Ksharp · Accepted solution]

title height=16pt "Var by Group";
proc sgplot data=test ; 
where Group is not missing;
   histogram Var1 / group=Group transparency=0.5 datalabelattrs=(size=16)  ;
   density Var1 / type=kernel group=Group ;
   keylegend /location=outside position=topright valueattrs=(size=14);
   xaxis display=(noline) labelattrs=(size=16) valueattrs=(size=16) ;
  yaxis display=(noline) labelattrs=(size=16) valueattrs=(size=16);

run;
quit;

[mateescu]

Perfect, thank you!