I see no inconsistency here as far as -duplicates-
is concerned.
-duplicates report- is telling you that, in terms of these
two variables, there are 1322511 observations, each of which is
unique. Thus, none are surplus, in the sense of repeating
information. That is, no duplicates, as -duplicates list- is also
telling you.
I cannot solve the -xtdes- mystery.
Nick
[email protected]
Stephen V. Burks
> Now I wonder if I am reporting a bug. As per Nick's suggestion
> (thanks, Nick) I looked up -duplicates- and ran it.
>
> Results:
> (1) using -duplicates report-
>
> . duplicates report DRVNUM CDATE
>
> Duplicates in terms of DRVNUM CDATE
> --------------------------------------
> copies | observations surplus
> ----------+---------------------------
> 1 | 1322511 0
> --------------------------------------
>
> (2) using -duplicates list-
>
> . duplicates list DRVNUM CDATE
> Duplicates in terms of DRVNUM CDATE
> (0 observations are duplicates)
>
>
> If I read this correctly, one version of -duplicates- says I have
> a single duplicate (the last case in the file), while the other
> says I have none. Also, -xtdes- still says that DRVNUM and
> CDATE do not uniquely identify cases.
>
> I looked in the data browser, and the last case is NOT a duplicate.
> While DRVNUM is the same as the case before, CDATE is not.
>
> If my interpretation of the output is correct then my best guess is
> that -xtdes- is wrong to say I do not have unique identification,
> and whatever bug causes this might also be behind the apparently
> incorrect response from -duplicates report-. In any event, it seems
> to be a problem that -duplicates list- and -xtdes- do not agree.
>
> Any better ideas?
>
> Thanks,
> Steve
>
> P.S. My version of Stata is 8.2 for WinXP, with current updates.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
