Re: st: Date format from string

[Nick Cox <njcoxstata@gmail.com>]

So, I gave you a gun and you shot yourself. Sorry about that. (I was
the first author of -destring-.)

Nick
njcoxstata@gmail.com


On 11 March 2014 15:42, Philipp Hagspiel <philipp.hagspiel@gmx.net> wrote:
> thank you very much for your response! I have found the problem. When I
> imported the data, I accidently -destring-ed and replaced the date variable
> so that I got numeric values. I imported it again and now it works!
>
> Thanks!
> Am 11.03.2014 11:19, schrieb Nick Cox:
>>
>> The error is "type mismatch". That means that Stata expects numeric
>> but sees string, or vice versa. In this case Stata expects string.
>>
>> So, the details of what's inside the date are as yet quite secondary.
>>
>> Compare the following examples:
>>
>> . di date(20012011, "DMY")
>> type mismatch
>> r(109);
>>
>> . di date("20012011", "DMY")
>> 18647
>>
>> . di date(string(20012011), "DMY")
>> .
>>
>> . di date(string(20012011, "%12.0f"), "DMY")
>> 18647
>>
>> In your case, try
>>
>>   gen eventdate = date(string(mydatevariable, "%12.0f"), "DMY")
>>
>> The third example above shows the default format of -string()- does
>> not help here. Why is that?
>>
>> . di string(20012011)
>> 2.00e+07
>>
>> -date()- can't make sense of that as a daily date.
>>
>> The word "format" is slippery here. None of your "formats" is in fact
>> a Stata format, although your meanings were all clear.
>>
>> Nick
>> njcoxstata@gmail.com
>>
>>
>> On 11 March 2014 10:07, Philipp Hagspiel <philipp.hagspiel@gmx.net> wrote:
>>
>>> I have imported and reshaped an excel file with dates as variables
>>> (observations after reshaping). They are now in the string format
>>> ddmmyyyy.
>>> They used to be in the format dd.mm.yyy in excel but I put a stub in
>>> front
>>> of every date, so I guess that is why it didn't keep the date format.
>>>
>>> When I know want to use the date function now, it tells me that there is
>>> a
>>> mismatch.
>>>
>>> . gen eventdate = date(mydatevariable, "DMY")
>>>      type mismatch
>>>      r(109);
>>>
>>> I have noticed that all dates that start with a "0" on the dd are
>>> converted
>>> into a d format.
>>> So: 01.01.2001 --> 1012011, 20.01.2011 --> 20012011
>>>
>>> Thus, I know have two formats: ddmmyyyy and dmmyyyy (whenever the day is
>>> between 01-09 it is now 1-9).
>>>
>>> How can I fix this?
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/