On the face of it, your data don't qualify as
panel data in Stata's sense. This is why -tsset-
doesn't play. Hoping to use -tsfill- is thus
out of the question. There isn't a fix for
this.
The only way you can get panel data is by
ensuring that there is at most one observation
for each combination of panel identifier,
which would require some kind of reduction,
e.g. a -collapse- to mean, minimum, maximum
or some single number per month.
Nick
[email protected]
Tobias Br�tsch
> I have a Problem with Tsset. I just tried with the link
> http://www.stata.com/support/faqs/data/repeatedtime.html
> But I cant find the solution.
>
> I have the following dataset:
> Firmname Rec. Rec_2 BNUM
> Id monthend
>
> |-------------------------------------------------------------
> --------------
> -----------
> 1. | ABB 4 2 ABN Amro
> 1 15705
> |
> 2. | ABB 3 5 BCV
> 2 15644
> |
> 3. | ABB 4 4 BNP Paribas
> 3 15613
> |
> 4. | ABB 4 4 BNP Paribas
> 3 15674
> |
> 5. | ABB 3 2 BNP Paribas
> 3 15764
> |
>
> |-------------------------------------------------------------
> --------------
> -----------
> 6. | ABB 2 2 BNP Paribas
> 3 15764
>
>
> It is a sample with stockrecommendations about 5 years. The variable
> monthend was generatet:
>
> - Gen daily = date(datum,"dmy"
> - Egen monthend = eomd(daily)
>
> The ID was generated:
>
> egen Id= group( titelname BNUM)
>
>
> Now i have to Tsset my data:
>
> - tsset Id monthend
>
> And then I get the "repeated time values within panel" - misstake.
> Of course there are repeated time values in my data, there are many
> observations about the same firm from the same broker in the
> same month, or
> sometimes on the same day.
>
>
> . duplicates report Id monthend gives me about 500 duplicates.
>
>
> How can I solve this problem? I cant delete all the
> duplicates and I have to
> analyse the data for each month. I have to tsset my data,
> because I want use
> tssfill after tsset.
*
* 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/
