Re: st: An econometric question

["Roy,Suryadipta" <suryadipta.roy@lawrence.edu>]

Dear Tak Wai,

Thanks! That probably is what I will be doing! I was 
trying to get some results on the average values of the 
variables ove the years, following previous studies. But 
with the original variables z, z1, and z2, the results are 
with the right signs and significant as well.

Thanks!
Suryadipta.

On Mon, 09 Apr 2007 13:45:31 -0400
 Tak-wai Chau <tchau@troi.cc.rochester.edu> wrote:

> Hi, Roy,
>
> What are your exact reasons why you don't want to put in 
> z1, z2 directly? How does it look like if you put z1 and 
> z2 instead of their country means and use 
> cluster(country)?
>
> Tak Wai
>
> Roy,Suryadipta wrote:
>
> > Hi everyone,
> >
> > Thank you so much for your comments- I am really 
> > learning a lot out of 
> > this. Regarding Austin's suggestion of using the 
> > cluster(id) option, its 
> > true that it makes the reults look similar to the 
> > collapse version in 
> > terms of significance. While I really want to take heart 
> > from Maarten's 
> > suggestion of not worrying about the significance (my 
> > signs are coming 
> > out great and exactly as predicted by the theoretical 
> > model!), I am 
> > worried as to how the referee's would take to it. 
> > Actually the 
> > explanatory variables are more like institutional 
> > measures which take a 
> > long time to change, but they do differ across countries 
> > a lot.
> >
> > In this regard, I was wondering if a cluster(region)- 
> > option instead of 
> > the cluster(country)- option would be appropriate. Any 
> > suggestion would 
> > be highly appreciated.
> >
> > Thanks!
> > Suryadipta.
> >
> >
> >
> > On Mon, 9 Apr 2007 12:46:00 -0400
> >  "Austin Nichols" <austinnichols@gmail.com> wrote:
> >
> > > Suryadipta Roy--
> > > If you use the 1500 observations, or 15 years of data on 
> > > 100
> > > countries, and you use the option cluster(country) on 
> > > your regression
> > > command, then you take account of dependence of the 
> > > errors within
> > > country over time, which allows the "effective sample 
> > > size" to be some
> > > number between 100 and 1500.  In the case where there is 
> > > really no new
> > > information across years, the cluster() option should 
> > > make the results
> > > look more like the -collapse-d version in terms of 
> > > statistical
> > > significance of coefficients.  The cluster() option is 
> > > advisable for
> > > the 1500 obs case even if you have a lot of 
> > > within-country variation
> > > because it is likely you have some kind of serial 
> > > correlation.
> > > Example:
> > > webuse grunfeld, clear
> > > egen mean_inv=mean(inv), by(comp)
> > > egen mean_k=mean(k), by(comp)
> > > reg mean_k mean_inv, nohe
> > > reg mean_k mean_inv, nohe cluster(comp)
> > > collapse k inv, by(comp)
> > > reg kstock inv, nohe
> > *
> > *   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/
*
*   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/