Dear Members,
I've a dataset which contain missing and non missing values. The "Spec" dataset contains the information about first dataset. I wanted to check missing values variable and wanted to create one more dataset which should contain the original dataset with one extra variable . I got stuck how i can concatenate variable names.
Output wanted:
a b c Missing_Details
1 . a1 B is missing
. 2 b2 A is missing
. . A is missing,B is missing,C is missing
1 2 c3
data test;
infile datalines;
input a b c $1.;
datalines;
1 . a
. 2 b
. .
1 2 c
;
run;
data Spec;
input Req name $ TYPE $;
datalines;
0 a NUM
0 b NUM
0 C CHAR
;
run;
proc sql;
select name into :name1 SEPARATED by ' ' from spec WHERE TYPE EQ 'NUM';
select name into :name_CH1 SEPARATED by ' ' from spec WHERE TYPE EQ 'CHAR';
quit;
OPTIONS MPRINT MLOGIC SYMBOLGEN;
DATA _NULL;
SET TEST;
ARRAY ARR[*] &name1;
array arr1[*] &name_ch1;
DO J=1 TO dim(ARR);
IF MISSING(ARR[J]) THEN MISS='MISSING';
END;
DO i=1 TO dim(arr1);
IF MISSING(ARR1[i]) THEN MISS='MISSING';
END;
RUN;
Thank you everyone in advance !!
Using what you've posted here a solution approach.
DATA want(drop=i j);
SET TEST;
ARRAY ARR[*] &name1;
array arr1[*] &name_ch1;
length Missing_Details $120;
DO J=1 TO dim(ARR);
IF MISSING(ARR[J]) THEN
Missing_Details=catx(',',Missing_Details,vname(ARR[J]));
END;
DO i=1 TO dim(arr1);
IF MISSING(ARR1[i]) THEN
Missing_Details=catx(',',Missing_Details,vname(ARR1[i]));
END;
RUN;
proc print data=want;
run;
N.B: The length of variable Missing_Details must be sufficient to hold all the variable names in case they are all missing (=32 characters for the var name plus 1 character for the comma per variable).
Hello,
data test;
infile datalines;
input a b c $1.;
array n _NUMERIC_;
array ch _CHARACTER_;
length miss $200;
do over n;
if missing(n) then miss=catx(' ', miss, vname(n), 'is missing');
end;
do over ch;
if missing(ch) then miss=catx(' ', miss, vname(ch), 'is missing');
end;
datalines;
1 . a
. 2 b
. . .
1 2 c
;
run;
Using what you've posted here a solution approach.
DATA want(drop=i j);
SET TEST;
ARRAY ARR[*] &name1;
array arr1[*] &name_ch1;
length Missing_Details $120;
DO J=1 TO dim(ARR);
IF MISSING(ARR[J]) THEN
Missing_Details=catx(',',Missing_Details,vname(ARR[J]));
END;
DO i=1 TO dim(arr1);
IF MISSING(ARR1[i]) THEN
Missing_Details=catx(',',Missing_Details,vname(ARR1[i]));
END;
RUN;
proc print data=want;
run;
N.B: The length of variable Missing_Details must be sufficient to hold all the variable names in case they are all missing (=32 characters for the var name plus 1 character for the comma per variable).
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.