I replied to Nisha Malhotra
> > I am facing a problem creating a binary variable. I have
> a large panel data
> > set. The three important variables are Year, Industry,
> Action taken by an
> > industry ( variable that takes the value 0 if no action
> was taken and 1 if one
> > action was taken and 2 if two actions were taken and so on )
> >
> > The data looks something like this
> > Year Industry Action
> > 90 A 0
> > 91 A 0
> > 92 A 2
> > 93 A 0
> > 90 B 1
> > 91 B 0
> > 92 B 0
> > 93 B 1
> >
> > I want to create a binary variable say BB, which takes
> the value 0 if no prior
> > action(in the previous years) has been taken by an
> industry and 1 if some
> > positive action has been taken by an industry in the
> previous years.
> >
> > Year Industry Action BB
> > 90 A 0 0
> > 91 A 0 0
> > 92 A 2 1
> > 93 A 0 1
> > 90 B 1 1
> > 91 B 0 1
> > 92 B 0 1
> > 93 B 1 1
>
> . tsset Industry year
>
> Note that if Industry is really string, you must -encode- it
> first and use the encoded version as panel variable.
>
> . gen byte BB = L.Action > 0 & L.Action < .
>
> Here the attitude over the first value of BB for each
> Industry differs from your example.
I think I misinterpreted this, as I read
"previous years" as "previous year".
If the precise definition is action in
any previous year, then
. bysort Industry (Year) : gen byte BB = sum(Action) > 0
Nick
[email protected]
*
* 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/
