creating categorical variable based on date (before/after)

[amigapomba]

hi all,

I imported an xlsx file into university addition using the import snippet. Date is in MM/DD/YYYY format. I assume SAS doesn't know that so I wrote a format statement to clarify. I then tried to create a categorical variable "reg" based on the date. The code I used is as follows:

data sc.date;
set sc2;
format date mmddyy10.;
run;

data sc.change;
set sc.date;
if date < 06/06/2014 then reg = "before";
else reg = "after";

run;

I'm not getting any errors, however all of my observations have been assigned "after", even though many of them have a date prior to 6/6/2014.

Thanks in advance for any help!

[Reeza · Accepted solution]

SAS is seeing, 06/06/2014 which is 6 divided by 6, divided by 2014. 

To specify a date, a specific format has to be used, the date9 format and enclosed in quotes and with a d at the end. 

if date < '06Jun2014'd then reg='Before';
else reg = 'After';

Date and time literals:

t   ->  time -> '08:00't

dt -> datetime

d  -> date

---

@amigapomba wrote:

hi all,

I imported an xlsx file into university addition using the import snippet. Date is in MM/DD/YYYY format. I assume SAS doesn't know that so I wrote a format statement to clarify. I then tried to create a categorical variable "reg" based on the date. The code I used is as follows:

data sc.date;
set sc2;
format date mmddyy10.;
run;

data sc.change;
set sc.date;
if date < 06/06/2014 then reg = "before";
else reg = "after";

run;

I'm not getting any errors, however all of my observations have been assigned "after", even though many of them have a date prior to 6/6/2014.

Thanks in advance for any help!

---

---

@amigapomba wrote:

hi all,

I imported an xlsx file into university addition using the import snippet. Date is in MM/DD/YYYY format. I assume SAS doesn't know that so I wrote a format statement to clarify. I then tried to create a categorical variable "reg" based on the date. The code I used is as follows:

data sc.date;
set sc2;
format date mmddyy10.;
run;

data sc.change;
set sc.date;
if date < 06/06/2014 then reg = "before";
else reg = "after";

run;

I'm not getting any errors, however all of my observations have been assigned "after", even though many of them have a date prior to 6/6/2014.

Thanks in advance for any help!

---

[amigapomba]

thank you!!