The FAQ cited discusses precisely this problem.
Please do read it.
Nick
[email protected]
Rashmi Shankar
> Hi, Nick: This worked last night when I tried it with a unit
> root test. But
> when I run the following loop, it doesn't work. The problem
> is that the
> dependent variable, bp_level, skips over some of my 56
> countries. The loop
> stops everytime it hits a country with no bp_level data.
>
>
> set more off
> tsset cnum obs_number , format(%tq)
>
> gen dbp_level = bp_level-bp_level[_n-1]
> local vari "lnpetrol lnchinaimp"
> foreach y of num 1/56 {
> foreach x of local vari {
> disp `y'
> dfuller `x' if cnum==`y' , trend lags(4) regress
> }
> }
>
> **1st differenced
> foreach y of num 1/56 {
> foreach x of local vari {
> gen d`x' = `x'-`x'[_n-1]
> disp `y'
> dfuller d`x' if cnum==`y' , trend lags(4) regress
> drop d`x'
> }
> }
> Quoting Nick Cox <[email protected]>:
>
> > Yes.
> >
> > http://www.stata.com/support/faqs/data/foreach.html
> >
> > may help. If not, you will need to tell us more
> > about your problem to get much better advice.
> >
*
* 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/
