Difference between Two dates

[danwarags]

Hello everyone,

I am trying to calculate the difference between two dates which are in different format as shown below. The start date is in the format yyyymmdd and end date is in the format ddmmmyy:hh:mm:ss

Sdate	Edate
20010130	01JUN01:09:01:00
20030524	04AUG03:16:10:00
20010421	01JUL01:08:15:00

The output dataset should show the difference between two dates in days and months. The differene should include end date in calculation. The output dataset should show as below. 

NoDays

123

73

72

Thank you in advance!

[Astounding]

For each variable, what does PROC CONTENTS tell you about it:  Is it numeric or character?  If numeric, does it have a format?

It doesn't matter what the variable looks like when you print it.  It matters what PROC CONTENTS reveals as the characteristics of the variable.

[Reeza · Accepted solution]

---

@danwarags wrote:

The output dataset should show the difference between two dates in days and months. 

How are you defining a 'month'? Your sample shows number of days and that's easy to calculate. 

Dates are stored as the number of days from Jan 1 1960. You can do math on them, ie subtract dates directly once they're SAS dates. To create SAS dates:

Convert datetime to date -> DATEPART(var)

Convert text to date -> INPUT( var, YYMMDD8.)

[danwarags]

Thanks Reeza. I really appreciate your help. The code worked. Given the start_date and End_date in the SAS format, how do I calculate the difference between the earliest start_date and late end_date within each ID. For example, I have a scenario as shown below:

ID	Start_date	End_date	Ddiff
1	1/1/2001	1/12/2001	17
1	1/2/2001	1/18/2001
1	1/6/2001	1/8/2001
2	3/4/2001	3/8/2001	4
3	2/4/2002	2/12/2002	20
3	2/10/2002	2/24/2002
3	2/14/2002	2/18/2002
4	3/15/2003	3/18/2003	11
4	3/20/2003	3/26/2003

I Should calculate Ddiff as mentioned in the above scenario. The Ddiff should be calculated in such a way that it should show the difference of early start_date and late end_date within each ID as shown above. 

Thank you!!

[Ksharp]

data have;
infile cards expandtabs truncover;
input ID	Start_date : mmddyy10.	End_date : mmddyy10.;
format Start_date End_date  mmddyy10.;
cards;
1	1/1/2001	1/12/2001	17
1	1/2/2001	1/18/2001	 
1	1/6/2001	1/8/2001	 
2	3/4/2001	3/8/2001	4
3	2/4/2002	2/12/2002	20
3	2/10/2002	2/24/2002	 
3	2/14/2002	2/18/2002	 
4	3/15/2003	3/18/2003	11
4	3/20/2003	3/26/2003	 
;
run;
proc sql;
create table want as
 select *,max(end_date)-min(start_date) as dif
  from have
   group by id;
quit;