Problem with data dates

[georgel]

Dear SAS users and experts, 

I would like to resolve the issue below for my datasets. 

I have the right format for the data Issue_date and Offer_datec but

I have tried to calculate the diffeence with meaningless results . Please see my file attached

The code 

data el.Data_test(keep=Issue_date Offer_date Offer_datec Dif_Issue_Offer);
    set Data_ipo4;
     * Dif_Issue_Offer=Listing_datec-Issue_date;
	   * Offer_datec=input(Offer_date,8.);
        *Offer_datec=Offer_datec-0;
        * format Offer_datec YYMMDDN8.;
	     Dif_Issue_Offer=Offer_datec-Issue_Date;
run;

Many thanks in advance,

George

[PaigeMiller · Accepted solution]

---

@georgel wrote:

Dear SAS users and experts, 

I would like to resolve the issue below for my datasets. 

I have the right format for the data Issue_date and Offer_datec but

---

Format is not the issue. Both unformatted variables must be number of days since 01JAN1960 in order for SAS to be able to treat them as dates. Your variable offer_datec is not on this scale. Thus any subtraction will yield gibberish.

data want;
    set el.Data_test;
    offer_daten=input(put(offer_datec,8.),yymmdd8.);
    Dif_Issue_Offer=Offer_daten-Issue_Date;
run;

Now Offer_dateN and Issue_Date are on the same scale, number of days since 01JAN1960, and the subtraction works.

[georgel]

Dear Miller,
Thanks a million for your quick respond.
It works perfectly

[Reeza]

offer_datec is a datetime, not date. 

Use datepart() to convert it to a date variable. 

---

@georgel wrote:

Dear SAS users and experts, 

I would like to resolve the issue below for my datasets. 

I have the right format for the data Issue_date and Offer_datec but

I have tried to calculate the diffeence with meaningless results . Please see my file attached

The code 

data el.Data_test(keep=Issue_date Offer_date Offer_datec Dif_Issue_Offer);
    set Data_ipo4;
     * Dif_Issue_Offer=Listing_datec-Issue_date;
	   * Offer_datec=input(Offer_date,8.);
        *Offer_datec=Offer_datec-0;
        * format Offer_datec YYMMDDN8.;
	     Dif_Issue_Offer=Offer_datec-Issue_Date;
run;

Many thanks in advance,

George

---

[Tom]

Those values are not datetime values either.  If they are they all from the year 1960.

---

@Reeza wrote:

offer_datec is a datetime, not date. 

Use datepart() to convert it to a date variable. 

---

@georgel wrote:

Dear SAS users and experts, 

I would like to resolve the issue below for my datasets. 

I have the right format for the data Issue_date and Offer_datec but

I have tried to calculate the diffeence with meaningless results . Please see my file attached

The code 

data el.Data_test(keep=Issue_date Offer_date Offer_datec Dif_Issue_Offer);
    set Data_ipo4;
     * Dif_Issue_Offer=Listing_datec-Issue_date;
	   * Offer_datec=input(Offer_date,8.);
        *Offer_datec=Offer_datec-0;
        * format Offer_datec YYMMDDN8.;
	     Dif_Issue_Offer=Offer_datec-Issue_Date;
run;

Many thanks in advance,

George

---

---

[Tom]

---

@georgel wrote:

Dear SAS users and experts, 

I would like to resolve the issue below for my datasets. 

I have the right format for the data Issue_date and Offer_datec but

---

NO. You have the WRONG format attached to one of those variables.

This is easy to see by looking at the actual VALUES that they contain.

proc means data="c:\downloads\data_test" min max mean;
  format _all_;
run;

Check those min/max values and see what date or datetime they represent.

data test;
  input date @@;
  put date=comma12. date :datetime19.  date date9. ;
cards;
 15000.00       20402.00
19880607.00    20200603.00
;

Results:

date=15,000 01JAN1960:04:10:00 25JAN2001
date=20,402 01JAN1960:05:40:02 10NOV2015
date=19,880,607 18AUG1960:02:23:27 *********
date=20,200,603 21AUG1960:19:16:43 *********

It looks like ISSUE_DATE has date values. 

But OFFER_DATEC has neither DATE nor DATETIME values.  It looks like it just has integers in the style YY,YYM,MDD.

So you need to convert the values stored in OFFER_DATEC into actual DATE values before you can use it as a date. Either in your calculations or to use with a display format that expects date values.

data el.Data_test ;
   set Data_ipo4(keep=ISSUE_DATE OFFER_DATEC);
   offer_datec = input(put(offer_datec,z8.),yymmdd8.);
   Dif_Issue_Offer=Offer_datec-Issue_Date;
   format issue_date offer_datec date9.;
run;

[ballardw]

Printing some observations from that data set shows this in the log:

220  proc print data=tmp1.data_test (obs=5);
221  run;

NOTE: There were 5 observations read from the data set TMP1.DATA_TEST.
NOTE: PROCEDURE PRINT used (Total process time):
      real time           0.01 seconds
      cpu time            0.00 seconds


ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.
ERROR: There was a problem with the format so BEST. was used.

The latest date that SAS will use is 31DEC20000 (yes roughly 18,000 years from now) which has a numeric value (number of days) of 6,589,335. Which is considerably smaller than the values of your Order_datec such as 20,141,002.

If 20141002 is supposed to be 02OCT2014 then you need to convert to an actual date. Something like this:

data example;
  x=20121005;
  datefromx = input(put(x,8. -L),yymmdd10.);
  format datefromx date9.;
run;