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: Econometrics question


From   Christopher Hajzler <chajzler@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Econometrics question
Date   Tue, 30 Mar 2010 08:21:09 +1300

This possibility was not intuitive for me, but giving it more thought
I can now imagine a scatter plot scenario with the time-period split
essentially chopping the group of points in half, which could result
in both subgroup coefficients even having even the opposite sign as
the pooled coefficient.  So Martin is right.

It might therefore be worthwhile plotting the residuals from the
regression that excludes GDP growth, distinguishing between each of
the sub-periods, to get a better idea of what might be driving this
result.

Cheers,
Chris

On Tue, Mar 30, 2010 at 8:06 AM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>
> <>
>
> Even w/o other covariates, the mystery lingers on :-)
>
>
> *******
> sysuse auto, clear
> reg price weight
> est store full
> reg price weight  in 1/`=_N/2'
> est store firsthalf
> reg price weight  in `=_N/2+1'/l
> est store secondhalf
> estimates table full firsthalf /*
>  */  secondhalf, se style(oneline) /*
>  */ stats(N)
> *******
>
>
> HTH
> Martin
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Christopher
> Hajzler
> Sent: Montag, 29. März 2010 21:02
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: Econometrics question
>
> In terms of the actual value of the estimated coefficient,my sense is
> that your intuition is correct in the case of a univariate regression.
>  With more than one independent variable, we need to take into account
> potential multicolinearity, and any potential impact changes in the
> average of these variables across subperiods have on the estimates.
>
> Try running the regressions again having dropped the other variables
> and see what you get.
>
> Cheers,
> Chris
>
>
> On Tue, Mar 30, 2010 at 7:50 AM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>>
>> <>
>>
>> The significance also depends on the number of observations in the
>> estimation sample, so the "weighted average" need not carry over to the
>> standard error estimation:
>>
>> ***
>> sysuse auto, clear
>> reg price weight length
>> est store full
>> reg price weight length in 1/`=_N/2'
>> est store firsthalf
>> reg price weight length in `=_N/2+1'/l
>> est store secondhalf
>> estimates table full firsthalf secondhalf, se style(oneline)
>> ***
>>
>>
>> HTH
>> Martin
>>
>>
>> -----Original Message-----
>> From: owner-statalist@hsphsun2.harvard.edu
>> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of kokootchke
>> Sent: Montag, 29. März 2010 20:39
>> To: statalist
>> Subject: st: Econometrics question
>>
>> Dear Stata users,
>>
>> I have a basic econometric question and I'm hoping you can help me out. I
> am
>> running a regression of bond spreads on various variables denoting
> domestic
>> economic conditions, and country fixed effects; I'm clustering my standard
>> errors by quarter, e.g.
>>
>> xi: regress LogSpread GDPgrowth DebtToGDP i.country, cluster(time)
>>
>> I have quarterly data for 40 different countries, although it's a very
>> unbalanced panel because the spread of the bond is for new bond issues and
> a
>> lot of countries don't issue new bonds every quarter. So, the data would
>> look something like this:
>>
>> Country   Time   Spread GDPgrowth
>> Argentina 1991q1 400    3.0
>> Argentina 1994q4 450    2.5
>> Argentina 2001q3 800    0.7
>> Brazil    1993q2 ...
>> Brazil    1993q4 ...
>> Brazil    1994q1 ...
>> Colombia ...
>> ...
>>
>> When I run a simple regression like the one above for the full sample, I
>> obtain a coefficient for GDPgrowth of -0.073***
>>
>> Then if I run this same regression for two separate subsamples for the
> years
>> 1991-1997 and 1998-2006, my coefficients for GDPgrowth are -0.056 and
> 0.009,
>> both insignificant.
>>
>> In my experience, the full sample coefficient would in general be some
> sort
>> of weighted average of the two coefficients obtained from subsample
>> regressions. So, I don't understand why this is not the case here...
>>
>> The number of observations in the two subsamples add up to the number of
>> observations in the full sample estimations.
>>
>> Any ideas?
>>
>> Thanks!
>> Adrian
>>
>>
>>
>>
>>
>> _________________________________________________________________
>> Hotmail: Trusted email with powerful SPAM protection.
>> http://clk.atdmt.com/GBL/go/210850553/direct/01/
>> *
>> *   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/
>
>
> *
> *   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/


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