Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: instrument variable issues in xtivreg


From   Srini <srini_vaduvoor@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: instrument variable issues in xtivreg
Date   Sat, 5 Apr 2008 07:40:05 -0700 (PDT)

Hi,

I am using smallpox mortality rates and smallpox vaccination rates from two provinces in British colonial India. I am attempting to show that 1) the vaccination had a negative and significant affect on mortality and 2) there are externalities on the mortality of the focal district as a result of the vaccinations in neighboring districts. 

I ran the following: 
xtivreg mort_rate (lag_vacc_rate = lag2_inpatient_rate lag3_inpatient_rate) if province == "A", fe first small

where I’m instrumenting the lagged vaccination rate with the 2nd and 3rd lags of inpatient rates in dispensaries. 

In order to see the effect of the vaccination rate in neighboring districts on the focal district’s mortality, the regression becomes:

xtivreg mort_rate  adj_dist_vacc_rate (lag_vacc_rate = lag2_inpatient_rate lag3_inpatient_rate) if province == "A", fe first small

However, in the first stage, STATA uses all three variables, adj_dist_vacc_rate, lag2_inpatient_rate and lag3_inpatient_rate, as instruments. I do not want this to occur. Is there a way I can avoid this issue?

Also, should I be using xtivreg2 or xtabond instead?

Thanks.



      ____________________________________________________________________________________
You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost.  
http://tc.deals.yahoo.com/tc/blockbuster/text5.com


*
*   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/



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