Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: DiD
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: DiD
Date
Tue, 29 Mar 2011 13:35:04 +0100
You seem confused on various levels.
The statement
. gen indicator = year > 2004
would make perfect sense to Stata. 2004 is just a constant, and need
not be present in your dataset as a value for -year-.
That said, creating a indicator for before and after sounds pointless,
as from your account there are just two distinct values in -year-.
That alone does not rule out regression, as regression is not limited
by the number of distinct values of any non-constant predictor:
otherwise using indicator variables of any kind would be invalid.
Nick
On Tue, Mar 29, 2011 at 1:25 PM, inggrid <[email protected]> wrote:
> Hi Federico,
>
> I only have data in 2000 and 2007. The reform took place in 2005 (before and after).
> So, I could not generate year>2004 etc.
> I checked my data. As your speculation, treat and post05treat have the same number of 1!
>
> Is there any suggestion how to run a did regression by using two different data points?
>
> I couldn't use a RD design, since I can't construct a threshold to determine the eligibility.
> --- On Tue, 3/29/11, [email protected] <[email protected]> wrote:
>> From: [email protected] <[email protected]>
>> Subject: Re: st: DiD
>> To: "inggrid" <[email protected]>
>> Cc: [email protected], [email protected]
>> Date: Tuesday, March 29, 2011, 7:06 PM
>> > gen post05 = year>2000
>> > gen treat_post05 = treat*post05
>>
>> > note: post05 omitted because of collinearity
>> > note: treat_post05 omitted because of collinearity
>>
>> May be "post05" has been omitted because it's always equal
>> to 1, and treat_post05 because it's always equal to "treat".
>> I guess you misspecified the condition of post05 (it
>> shouldn't be "year>2000", but "year>2004" or
>> "year>2005").
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/