Re: st: RE: growth rate

[Nick Cox <njcoxstata@gmail.com>]

These methods ignore the fact that Lynn's gap varies between 5 years
and 2 years. That's neither necessary nor accurate. See my post for a
constructive alternative.

Nick

On Wed, Oct 10, 2012 at 2:39 PM, Swanquist, Quinn Thomas
<qswanqui@utk.edu> wrote:
> If you must use xtset to do this, you can create a new variable to identify the year as follows:
>
> sort id year
> by id: gen yearid = _n
> xtset id yearid
> gen growth=D.loggdp
>
> Alternatively you could use:
>
> sort id year
> by id: gen growth = loggdp - loggdp[_n-1]
>
> Of course your growth rate is based on unbalanced panel data so you'll need to consider that when using this variable for its intended purpose.
>
> Quinn T. Swanquist
> qswanqui@utk.edu
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Lynn Lee
> Sent: Wednesday, October 10, 2012 9:25 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: growth rate
>
> Nick,
>
> My panel data has 4 years for each country,2007,2002,1997,1995. The gap is not same. Therefore, I do not know how to assign value to -delta()- in -xtset-. I figured out why  Stata11 generated all missing values for growth rate. But the year gap is unbalanced. I tried -xtset id year,delta(5)-,
> Stata11 told me "time values with period less than delta() found.
>
> Now, I replace "1995" with "1992" artificially, then I got the growth rate.
>
> But is there any direct way (not need to replace the year artificially) which can generates growth rate for unbalanced year gap panel data?
>
> I appreciate for your help.
>
> Best Regards,
> Lynn Lee
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
> Sent: Wednesday, October 10, 2012 8:49 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: growth rate
>
> -help xtset- explains about setting by panel and year. Once you have done that correctly
>
> gen growth=D.loggdp
>
> is sufficient.
>
>
> On Wed, Oct 10, 2012 at 1:41 PM, Lynn Lee <lynn09v@gmail.com> wrote:
>
>> I use -bysort id: gen growth=D.loggdp- to generate growth rate of GDP
>> in panel data. I also wrote -xtset year- before this command, but
>> Stata11 told me "time variable not set". I am assuming this command is
>> only for time-series data. What about panel? What is the command used
>> in panel to generate growth rate?

*
*   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/