Thanks to both Nick and Oleksandr. In reply to Nick's
question I am not sure why it doesn't work.
Reka
----- Original Message -----
From: Nick Cox <[email protected]>
Date: Monday, May 5, 2003 5:14 am
Subject: Re: st: creating lag variables
> Reka Sundaram-Stukel
>
> > > I am trying to create a lag y for data that looks like
> > the following:
> > >
> > > id type choice Year lag_choice
> > > 1 a 1 1 -
> > > 1 b 0 1 -
> > > 1 c 0 1 -
> > > 1 d 0 1 -
> > > 1 a 0 2 1
> > > 1 b 0 2 0
> > > 1 c 1 2 0
> > > 1 d 0 2 0
> > >
> > > how do I generate lag_choice. I have tried
> > >
> > > by id type: gen lag_choice=choice[_n-1] and various
> > combinations of this
> > > quite unsuccessfully. Help please
>
> Oleksandr Talavera, BC
>
> > Try the following:
> > *----------
> > egen type1 = group(type)
> > g newid = 10^6*type1+id
> > tsset newid year
> > g lag_choice=L.choice
> > sort id year type
>
> Oleksandr's approach works. What's not
> clear is why Reka reported failure.
>
> bysort id type (year) : gen lag_choice = choice[_n-1]
>
> works with this example.
>
> The nuance separating this from an
> approach based on -tsset- is that [_n-1]
> looks backwards across a gap, but L. doesn't.
>
> 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/
>
*
* 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/
