Re: st: tsset for panel data with many variables

[Tetsuji Tanaka <tetsujitanaka.w@gmail.com>]

Thanks again, Nick, for the quick reply.
I will try to change the expression for dates and will get the STATA
understand it properly.

TJ







On Mon, 4 Nov 2013 21:09:35 +0000
Nick Cox <njcoxstata@gmail.com> wrote
> It's really not clear what you are doing. Which command produces that
> error message?
> 
> If you are telling Stata that you have monthly dates which are
> integers like 200610, to Stata they are several thousand years into
> the future. For Stata month 0 was January 1960 and this is month 646:
> 
> . di %tm 0
>  1960m1
> 
> . di %tm 646
> 2013m11
> 
> Conversely, if you are telling Stata something else, what is it?
> 
>  It is futile deciding on your own date convention and hoping that
> Stata will understand it. There is only one rule: follow Stata's date
> conventions, or invent your own completely separate system.
> 
> In the long run it is a waste of time _not_ to read -help dates and
> times- until what you need to know is clearly understood.
> 
> Please show us the results of
> 
> . describe case_id monthly_var
> 
> . list case_id monthly_var in 1/20
> 
> The request that you read
> http://www.stata.com/support/faqs/resources/statalist-faq/#spell still
> stands.
> 
> Nick
> njcoxstata@gmail.com
> 
> On 4 November 2013 20:43, Tetsuji Tanaka <tetsujitanaka.w@gmail.com> wrote:
> 
> > Dear all, and thank you for the response, Nick.
> >
> > I am trying that the STATA recognise my panel data as a panel data.
> > I made an id variable "case_id" to identify individual data (to avoid data duplication).
> > My monthly data variable is written like "200601" "200612" meaning January and December in 2006, respectively.
> >
> > format monthly_var %tm
> > tsset case_id monthly_var
> >
> > And I got an error message
> >
> > =exp required
> > r(100);
> >
> > I don't understand what is wrong.
> > Please give me ideas to overcome these problems.
> 
> Nick Cox <njcoxstata@gmail.com> wrote
> 
> >> You don't give your exact command. Presumably you tried something of the form
> >>
> >> tsset panelid time
> >>
> >> If you omitted a panel identifier variable, that will be problem: such
> >> a variable is essential. Otherwise see
> >>
> >> http://www.stata.com/support/faqs/data-management/repeated-time-values/
> >>
> >> http://www.stata.com/support/faqs/resources/statalist-faq/#spell
> 
> On 3 November 2013 10:52, Tetsuji Tanaka <tetsujitanaka.w@gmail.com> wrote:
> 
> >> > I am an economist and am trying to make the STATA  recognise my panel
> >> > data of prices as a panel data using the tsset.
> >> > However, the dataset has many variables and the STATA gave me an error
> >> > message "repeated time values within panel".
> >> >
> >> > My data has monthly time variable, price, region, shop type, shop size
> >> > etc.
> >> >
> >> > Please let me know about how to deal with the problem if you had similar
> >> > experiences.
> *
> *   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/



-------------------------------------------- 
Tetsuji Tanaka 
PhD, Research Fellow, Lifelong Health Theme
Rowett Institute of Nutrition & Health
University of Aberdeen
Greenburn Road, Bucksburn
Aberdeen, Aberdeenshire
AB21 9SB
United Kingdom
Email: tetsujitanaka.w@gmail.com
Tel. +44 (0)1224 43 8747 (Rowett) or 
       +44 (0)1224 43 7128 (Polwarth)  
Mobile: +44 (0)750 361 8172

*
*   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/