Statalist The Stata Listserver

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

st: ssm wrapper, problem with LRtest for rho? (redux)

From   Jeff Svien <>
Subject   st: ssm wrapper, problem with LRtest for rho? (redux)
Date   Sat, 2 Dec 2006 11:33:33 -0800 (PST)

I'm running into the same problem reported below (Mon,
2 Oct 2006).  

When I try running SSM on the example in the help file
I get an r(2000) error, even though there appears to
be observations in the simulated data set.

Additionally, the command runs EXTREMELY slowly (as in
multiple hours), which is surprising given that I'm
running it over UNIX on Stata SE.  

Generically, the command that I'm running looks like

xi: ssm y1 x1 x2 i.x3 i.x4 x5 i.x6 i.x7, s(x1 = x8 x9
x10 x11 x5) adapt q(16) family(bin) link(logit)

Any suggestions would be much appreciated!


Hi all, 
   When using the new ssm wrapper to fit a logit model
with a single endogenous dummy, I'm having trouble
with the likelihood ratio test for rho==0. The model
converges, but produces strange results for that test:

. ssm sexgrat52 age harassd puberty11 firstvi ktouch
usborn2 [pweight=rweight] if xm==1, s(harassd =
puberty11 firstvi ktouch usborn2 edimp ) adapt q(16)
family(binom) link(logit)

    |  Coef.   Std. Err  z    P>|z|     [95% CI]
rho |.6419742 .0149583 42.92  0.000 .5999316 .6639211
Likelihood ratio test for rho=0: chi2(1)= 0.00
Prob>=chi2 = 1.000

In general, no matter what model I run with ssm, chi2
is *always* = 0.00. Any thoughts? Thanks. 
                      -Aniruddha Das.

Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
*   For searches and help try:

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