Hello,
I am experiencing exactly the same problems as those in Wiji Arulampalam's
posting (see email subject title). I am trying to estimate a RE probit (
with a lagged dependent variable) and I am using Wooldridge's 2005 ( Journal
of Applied Econometrics) solution to the initial conditions problem that
still permits to estimate the model using the xtprobit command. However,
while I can actually get proper estimations with convergent values ( eg full
rank variance covariance matrix) and standard errors for two of my models (
male1997-2002, female1991-1996) this does not seem to be the case with my
(male1996-2002 and female 1997-2002) models. The maximum likelihood
estimation for the male sample 1991-1996 stops at rho=0.1 when fitting the
the full model and then the results are such that no standard errors are
reported for lnsig2u , sigma_u and rho ( although s.es are reported for all
the rest) Note that, the balanced panels do not seem to share any markedly
different characteristics ( eg descriptive stats of explanatory variables
are reasonably similar, correlation coefficients between variables are also
genarally low and similar for both time periods with regard to both male and
female sub samples).
For the female 1997-2002 sample though, the maximisation process does
proceed normally to rho=0.2 but then the estimates are again such that no
s.es are given for rho lnsig2u, sigma_u ( though in this case I get
something like :
Likelihood-ratio test of rho=0: chibar2(01) = 113.35 Prob >= chibar2 =
0.000 for the LR test whereas in the previous one when it does not proceed
to rho=0.2 I get a Prob >= chibar2 = 1.000.
I have also tried to use the random effects regression model (12.6, pp.
226-229 :Maximum Likelihood Estimation with Stata (2nd edition) i.2 the
myrereg model but to my understanding this estimation command does not do
anything for the initial conditions problem. The only thing it does is that
it sets some initial values for the intercept in the constant only model and
the initial values of the variances are taken form the variance components
from an one-way ANOVA of the constant only model. And then again if I use
the myrereg estimation command I get
Very perverse results (lns_e has a z value: -220.78 ). The only way this
gives somewhat rational results - and still the z value on the lns_e is
quite high- is if I combine it with Wooldridge's approach ( eg include an
intial value of the lagged dependent variable in the eqn and also time
interaction dummies of variables that change over time) and I am not even
sure if this can be combined with the myrereg command.
Thus, none of these seems to work for me.
I was thinking of following Heckmans ( 1981) solution to the problem of the
initial condtitions ( in Manski and MacFadden - Structural Analysis of
Discrete Data and Econometric Applications but I have not got a clue as to
how to implement this practically. Would be most grateful if you could help
and I do apologise for the length of this message.
George
*
* 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/
