I see no bug here, but only what -duplicates- was
intended to do.
-duplicates- reports that you have 128 observations,
which are all unique (that is, there is just 1 copy of each
observation).
If -duplicates- found any duplicates,
they would exist as 2 or more copies of each other.
There is more explanation at [R] duplicates.
So -duplicates report- and -duplicates list- are
reporting consistent results.
Your problem is perhaps best investigated by asking
. duplicates list id year
Nick
[email protected]
Joana Quina
> Thanks for replying. I am sorry for the mix up - I was using "width"
> with "xtdes" and not with "duplicates".
>
> Here is what I typed and the output:
>
> . duplicates report
>
> Duplicates in terms of all variables
>
> --------------------------------------
> copies | observations surplus
> ----------+---------------------------
> 1 | 128 0
> --------------------------------------
>
> . duplicates list
>
> Duplicates in terms of all variables
>
> (0 observations are duplicates)
>
> . iis id
>
> . tis year
>
> . xtdes
>
> id: 1, 2, ..., 32
> n = 32
> year: 1, 2, ..., 4
> T = 4
> Delta(year) = 1; (4-1)+1 = 4
> (id*year does not uniquely identify observations)
>
> Distribution of T_i: min 5% 25% 50%
> 75% 95% max
> 2 4 4 4
> 4 5 5
>
> Freq. Percent Cum. | Pattern
> ---------------------------+---------
> 29 90.63 90.63 | 1111
> 1 3.13 93.75 | 1..1
> 1 3.13 96.88 | 1121
> 1 3.13 100.00 | 1211
> ---------------------------+---------
> 32 100.00 | XXXX
>
> . xtdes , width(200)
> option width() not allowed
> r(198);
> On 05/07/05, Nick Cox <[email protected]> wrote:
> > I am interested in your assertion that -duplicates-
> > produces an incorrect result. But to say anything
> > about it would require a sight of exactly what you
> > typed and exactly what results you got, and/or a copy of the
> > data, which you might care to send to me privately.
> >
> > I wasn't aware that -duplicates- had a -width()- option.
> >
> > Nick
> > [email protected]
> >
> > Joana Quina
> >
> > > My problem is that -xtdes- says my two key variables do
> not together
> > > uniquely identify the cases in my large panel data set. I
> am working
> > > with a small panel (N=32 and T=4).
> > >
> > > I have no duplicates, though "duplicates report" wrongly
> says I have
> > > one dupliacate observation. Curiously, "duplicates list" correctly
> > > says I have none.
> > > This appears to have happened to the only other query I
> could find on
> > > this subject. I have tried changing width, but stata says "option
> > > width() not allowed".
> > >
> > > I don't know what else to do. It seems quite a ridiculous
> problem for
> > > such a small data set! Thank you in advance for any pointers.
> > >
> > > Joana
> > >
> > > PS - I have Stata 8
*
* 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/
