LR test and rho statistics in bioprobit

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

LR test and rho statistics in bioprobit

This post has NOT been accepted by the mailing list yet.
Dear Statalist users,

I run a bioprobit model with Stata11 and I wondered if anybody can help me with an explanation for the estimated rho and LR test. When I try to run the following code

bioprobit (y1 = x1 x2 x3) (y2 = x1 x2), end
*Where y1 is my endogenous regressor and is instrumented using x3

I get the following results
rho : Coef = -.1001432, Std.Err. = .1201227
LR test of indep. eqns. : chi2(1) = 5.01,   Prob > chi2 = 0.0253

I think LR test in bioprobit tells us the independence of equations hypothesis (that is, rho=0) and "Prob > chi2(1)= 0.0253" indicates the null hypothesis is rejected and correlation might exist. However, on the contrary, the statistics of rho indicates it is not significant. In this case, how do I interpret the correlation between equations?

I would greatly appreciate any assistance which you might be able to provide.

Thanks much