[prev in list] [next in list] [prev in thread] [next in thread] 

List:       sas-l
Subject:    Re: Proc Transpose. Is it possible?
From:       Adam Hendricks <ahendrix () AA ! NET>
Date:       1997-08-30 18:00:40
[Download RAW message or body]

In article <34078424.21E7@interlog.com>, rcc@interlog.com says...
>
>Hi.
>    I have the following dataset:
>
>  OBS     TRANS     DATE     CYCLE1     CYCLE2     CYCLE3
>
>   1      00001   01JAN1996  JAN1996    FEB1996    MAR1996
>   2      00002   01FEB1996  FEB1996    MAR1996    APR1996
>   3      00003   01MAR1996  MAR1996    APR1996    MAY1996
>
>
>
>  I would like the final dataset to look like this:
>
>  OBS     TRANS     DATE     JAN1996  FEB1996  MAR1996  APR1996  MAY1996
>
>   1      00001   01JAN1996  CYCLE1   CYCLE2   CYCLE3
>   2      00002   01FEB1996           CYCLE1   CYCLE2   CYCLE3
>   3      00003   01MAR1996                    CYCLE1   CYCLE2   CYCLE3
>
>
>  I would like this dataset cause then I can quickly find which cycle a
>TRANS is at, on any particular date. I've tried PROCC TRANSPOSE but I
>can't seem to get it right. Also I want to avoid a messy dataset step
>routine. Any quick and easy solution? Any help is greatly appreciated.
>Thanks in advance.
>
>    Rob.

Try:

data pre;
  format trans z5. date cycle1-cycle3 monyy7.;
  informat date cycle1-cycle3 date9.;
  input trans date cycle1 cycle2 cycle3;
  cards;
1 01JAN1996 01JAN1996 01FEB1996 01MAR1996
2 01FEB1996 01FEB1996 01MAR1996 01APR1996
3 01MAR1996 01MAR1996 01APR1996 01MAY1996
;
run;

proc sort data=pre;
  by trans date;
run;

proc print data=pre;
  title 'Before';
run;

proc transpose data=pre out=temp;
  by trans date;
  var cycle1-cycle3;
run;

proc transpose data=temp out=post(drop=_name_ _label_);
  by trans date;
  id col1;
  var _name_;
run;

proc print data=post;
  title 'After';
run;

Adam Hendricks
TAD Staffing (Boeing)
Seattle, WA

[prev in list] [next in list] [prev in thread] [next in thread]