Converting to a date from YMD

[Stephan Molekski]

Hello everyone,




I have a data set with the variable date in it and it looks like this

date

19990106

20000425

19990930

...

it is in this format currently:

                         storage                     display     value
          variable       name   type          format     label      variable label 
          date            long    %12.0g

How can I convert it to a normal date, so I can substract other dates from it? Because I have a second date variable in this format, which I think should be correct?: (And it looks like 13sept1999 in the editor)

                                         storage            display    value
                variable            name   type    format     label      variable label
                 event_date      int     %td                                      event_date

I tried the following but I got an error:

 . gen edate2 = date(date, "ymd")
 type mismatch r(109);

I'd be really thankful for any ideas, Thanks!

[Nick Cox]

Look at the help for date() which starts

date(s1,s2[,Y])
       Description:  the e_d date (days since 01jan1960) corresponding to s1 based on s2 and Y

                     s1 contains the date, recorded as a string, in virtually any format.

So date() expects a string and you are feeding it a numeric variable: hence the error message, which should not be surprising.

This will work, but note further that you need YMD as argument not ymd:

clear 
input long date
19990106
20000425
19990930
end 


gen wanted = daily(string(date, "%8.0f"), "YMD") 
format date %8.0f 
format wanted %td 

list 

     +----------------------+
     |     date      wanted |
     |----------------------|
  1. | 19990106   06jan1999 |
  2. | 20000425   25apr2000 |
  3. | 19990930   30sep1999 |
     +----------------------+

[Nick Cox]

Cross-posted at https://www.reddit.com/r/stata/comme...date_from_ymd/

Please note our policy on cross-posting, which is that you are asked to tell us about it. https://www.statalist.org/forums/help#crossposting