[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: AW: Probit Model with Instrumental variables

From   "Martin Weiss" <>
To   <>
Subject   st: AW: Probit Model with Instrumental variables
Date   Mon, 27 Apr 2009 09:18:51 +0200


Well, it does run for "4.5", and as you -set seed- you should get the same
behavior. Note I alerted you to the fact that "4" was just a guess on my
part. What you can do is sneak a 

proportion y

into your -program- after the last -generate- and let it run to see what
kind of distribution of "y" you get from the statement. For the case of "4",
it is 98% vs 2%, which may or may not be what you want.

Also note that an error is easier to diagnose when you run the -prog- by
typing "endoprob" than when -simulate- executes it.


-----Ursprüngliche Nachricht-----
[] Im Auftrag von Sachin
Gesendet: Montag, 27. April 2009 01:42
Betreff: st: Probit Model with Instrumental variables

Dear All
Well I did find why the results were contrary to the results published
by Buse. I did notice one thind that Martin help me do but cannot
figure out why and hope someone could help me
When I use the statement -generate y = 0.5 + 2*x1 + x2hat + mu + u>4-
the y generated is 1 or 0 with "4' from which on we want the dependent
to be 1. Unfortunately when I change the value from "4" to anything
above then it does not work giving an error message that
"May not drop an endogenous regressor".
This has left me puzzled and stumped. Any help would be greatly appreciated.
global numobs 500             // sample size N
global numsims "500"         // number of simulations
set seed 123456789

capture program drop endoprob

program endoprob // , rclass
version 10.1
drop _all
set obs $numobs
   generate u = rnormal(0)
   generate mu = rnormal(0)
	generate v = rnormal(0)
   //generate x = rnormal(0)
   generate z1 = rnormal(0)   // 4 Instruements
   generate z2 = rnormal(0)
   generate z3 = rnormal(0)
   generate z4 = rnormal(0)
   // Also be written *drawnorm u mu x1 z1 z2 z3 z4
   generate a = 0.4*u
   generate x = 0.5 + 0.1*z1 + a
   regress x  z1 // endogenous regressor with four instruments
   predict px

  generate y = 0.5 + px + mu + u > 4    //Reduced Form Equation
  ivprobit y (x= z1 )
  predict py
simulate _b _se y py , ///
   reps($numsims): endoprob

mean x* y*
*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index