st: RE: May not drop an endogenous regressor

["Martin Weiss" <martin.weiss1@gmx.de>]

<>

Try to run the first stage regression only and see whether that works out.
Probably you will see some message warning of collinearity.


HTH
Martin

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Mafalda Sampaio
Sent: Dienstag, 23. Februar 2010 20:17
To: statalist@hsphsun2.harvard.edu
Subject: st: May not drop an endogenous regressor

Hi all,

 

I was trying to do the following regression:

 

eststo: xi: ivprobit m2_FP_chosen_AIDS $controls_01 $controls_partner_01
$socialdistance m2_AIDSnpartner (i.m2_worry_bin
i.m2_worry_bin*i.m2_modernFP_nwp = i.m1_worry_bin
i.m1_worry_bin*i.m2_modernFP_nwp)  if  m2_network==1 & m2_gender==0, cluster
(respid)

 

 

 

Where I have one endogenous regressor m2_worry_bin that is being
instrumented by m1_worry_bin. As I also want to evaluate the impact of the
interact of between the endogenous variable and another variable called
m2_modernFP_nwp I include both variables, with the corresponding two
instruments. However STATA provides the following error message:

 

 

. eststo: xi: ivprobit m2_FP_chosen_AIDS $controls_01 $controls_partner_01
$socialdistance m2_AID

> Snpartner (i.m2_worry_bin i.m2_worry_bin*i.m2_modernFP_nwp =
i.m1_worry_bin i.m1_worry_bin*i.m2

> _modernFP_nwp)  if  m2_network==1 & m2_gender==0, cluster (respid)

i.m2_agecat3      _Im2_agecat_2-5     (naturally coded; _Im2_agecat_2
omitted)

i.m2_educ         _Im2_educ_0-2       (naturally coded; _Im2_educ_0 omitted)

i.m2_marital      _Im2_marita_1-4     (naturally coded; _Im2_marita_1
omitted)

i.m2_location~p   _Im2_locati_1-6     (naturally coded; _Im2_locati_1
omitted)

i.m2_educatio~p   _Im2_educat_0-2     (naturally coded; _Im2_educat_0
omitted)

i.m2_pa~_friend   _Im2_partne_0-1     (naturally coded; _Im2_partne_0
omitted)

i.m2_partner~ve   _Im2_partnea0-1     (naturally coded; _Im2_partnea0
omitted)

i.m2_partner~ce   _Im2_partneb0-1     (naturally coded; _Im2_partneb0
omitted)

i.m2_strength~p   _Im2_streng_1-4     (naturally coded; _Im2_streng_1
omitted)

i.m2_worry_bin    _Im2_worry__0-1     (naturally coded; _Im2_worry__0
omitted)

i.m2_modernFP~p   _Im2_modern_0-1     (naturally coded; _Im2_modern_0
omitted)

i.m2~in*i.m2_..   _Im2_Xm2__#_#       (coded as above)

i.m1_worry_bin    _Im1_worry__0-1     (naturally coded; _Im1_worry__0
omitted)

i.m1~in*i.m2_..   _Im1_Xm2__#_#       (coded as above)

may not drop an endogenous regressor

r(498);

 

 

Any idea what is going on and how to overcome this?

 

Thanks!

 

Best,

 

Mafalda

*
*   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/