Re: st: RE: RE: Controlling for associated observations

[Elena Quercioli <elena.liquorice@gmail.com>]

Dear Statalist,

I need to formulate in STATA a panel regression in which I tie down
pairs of observations to have the same fixed effects.  In my case, I
am using a random effects MLE estimation procedure for my project and
I have a set of data for counties along time zone lines. I want to
estimate the effect on the observations of landing on the west side of
the time zone line. Of course, I also have other controls. But I want
to think of the counties that are adjacent as pairs, and thus
estimating a 'common fixed effects' through the years.
Any help would be greatly appreciated. Thank you.
Elena
P.S. : Intuitively this should reduce the number of parameters to
investigate and strengthen the results.

On Fri, Mar 11, 2011 at 4:16 AM, Jan Bryla <JBR@finansraadet.dk> wrote:
> Dear Elena
> This may be slightly advanced if you are new, but I would go about your problem using the -psmatch2- routine by Leuven and Sianesi, using nearest neighbour, no replacement, and a scaled latitude (to between 0 and 1) as the propensity score.
>
> The automatically generated additional variables _id and _n1 should provide you with the information you need. I took the liberty of writing a sample code to illustrate my point:
>
> //
> Drawnorm outcome, n(100)
> Gen pscore=uniform()
>  drawnorm help, n(100)
> gen treat=0
> replace treat=1 if help>0
>
> psmatch2 treat, outcome(outcome) pscore(pscore) noreplacement neighbor(1) caliper(1)
> //
>
> Then for each treated (say, to the east of the time zone) the _n1 gives you the _id of the closest county in terms of latitude of a non-treated (ie. To the west of the time zone).
>
>
> Hope this is what you are looking for
>
> Jan Bryla
> The Danish Bankers Association
>
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Elena Quercioli
> Sent: 10. marts 2011 22:30
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: Controlling for associated observations
>
> Dear Statalist members,
>
> I have a set of data from counties at different latitudes along the
> time zone lines in the USA. Counties lie to the east and to the west
> of each line. For my project I need to insert a (right hand side
> control) variable in my (panel) regression pairing up counties on
> opposite sides of a time zone line, by latitude or perhaps pairing
> "adjacent ones".
> Does anyone know how I could do this in Stata? I am relatively new to
> this and not really able to manage Stata beyond basic stuff.
>
> Thank you very much in advance, Elena
> *
> *   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/
>
> *
> *   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/
>

*
*   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/