Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Difference Model


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Difference Model
Date   Wed, 2 Nov 2011 08:21:20 +0000

That establishes that my comments on selecting observations may be
pertinent. But on the other hand if you look at only those
observations, you lose any dynamics and appear to assume instantaneous
effects within the time resolution of the data.

But you need economists working in this area to comment on which
models might make sense -- and with annual data??? I am not an
economist, so bail out at this point.

Nick

On Wed, Nov 2, 2011 at 8:00 AM, Vikram Finavker
<vikramfinavker@gmail.com> wrote:
> Hi,
>
> Sorry for not explaining why I want to use the model. Actually, I want
> to examine the effects of hedging on risk measures when a firm either
> starts hedging or stops hedging i.e. when h1 changes from 0 to 1 or
> from 1 to 0. Hope this makes it clear now.
>
> Regards,
>
> Vikram Finavker
>
> On 1 Nov 2011, at 11:48 PM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
>
>> I don't see how you think we can help. There is no context here and no explanation of why any model might make sense for whatever is your research problem.
>>
>> If you are saying that you want to select times at which -h1- differs from its previous value then
>>
>> ... if h1 != h1[_n-1]
>>
>> selects such observations.
>>
>> Nick
>> n.j.cox@durham.ac.uk
>>
>> vikramfinavker
>>
>> I have a data like this...
>>
>> year  firmid h1 x1 x2    x3 x4  x5
>> 1998     1      0  1    10
>> 1999     1      0  1    11
>> 2000     1      0  1    12
>> 2001     1      0  1    12
>> 2002     1      1  1    13
>> 2003     1      1  1    18
>> 2004     1      1  1    17
>> 2005     1      1  1    16
>> 2006     1      1  1    15
>> 2007     1      1  1    14
>> 1998     2      0  1    10
>> 1999     2      0  1    11
>> 2000     2      0  1    12
>> 2001     2      0  1    12
>> 2002     2      0  1    13
>> 2003     2      0  1    18
>> 2004     2      1  1    17
>> 2005     2      1  1    16
>> 2006     2      1  1    15
>> 2007     2      1  1    14
>>
>> Now i want to run regression using following way;
>> reg x3 h1 x2
>>
>> but i want to use only those observations where variables "h1" changes to
>> either "1" or "0". can i use first difference model.? Does it help me in
>> this case? can i run regression like:
>>
>> reg x3 d.h1 x2
>>
>> or do i have to run it like this
>>
>> reg d.x3 d.h1 d.x2
>>
>> Please let me know if you can help.
>>

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index