st: RE: non convergence when running Heckman selection model

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

<>

The manual [R], p. 654, reports similar problems, and implies that the
-twostep- version may be more stable during estimation.


HTH
Martin


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Thu Phuong Truong
Sent: Donnerstag, 4. März 2010 04:04
To: statalist@hsphsun2.harvard.edu
Subject: st: non convergence when running Heckman selection model

Dear all,

I have tried to run this below Heckman selection model. However, it seems
like my regression does not converge. Could anyone kindly help?

. heckman precision reform analyst herfindahl5 factor1 badnews goodnews
lnabsec_w lnassets_w cros
> slisted lnmb_w mefevent horizon, select(forecast1=reform analyst
herfindahl5 factor1 ecsign lna
> bsec_w lnassets_w crosslisted lnmb_w)

Iteration 0:   log likelihood = -1528.4141
Iteration 1:   log likelihood =  -1391.554  (not concave)
Iteration 2:   log likelihood =  -1346.221  (not concave)
Iteration 3:   log likelihood = -1338.1855  (not concave)
Iteration 4:   log likelihood = -1335.3504  (not concave)
Iteration 5:   log likelihood = -1332.2101  (not concave)
Iteration 6:   log likelihood = -1322.7847  (not concave)
Iteration 7:   log likelihood = -1320.0481  (not concave)
Iteration 8:   log likelihood = -1316.2955  (not concave)
Iteration 9:   log likelihood = -1314.2313  (not concave)
Iteration 10:  log likelihood = -1310.3296  (not concave)
Iteration 11:  log likelihood = -1307.7817  (not concave)
Iteration 12:  log likelihood = -1306.5784  (not concave)
Iteration 13:  log likelihood = -1305.1797  (not concave)
Iteration 14:  log likelihood = -1304.4682  (not concave)
Iteration 15:  log likelihood = -1303.2272  (not concave)
Iteration 16:  log likelihood = -1302.3122  (not concave)
Iteration 17:  log likelihood = -1301.7086  (not concave)
Iteration 18:  log likelihood =  -1300.982  (not concave)
Iteration 19:  log likelihood = -1300.5817  (not concave)
Iteration 20:  log likelihood = -1298.9577  (not concave)
Iteration 21:  log likelihood = -1298.0653  (not concave)
Iteration 22:  log likelihood = -1297.6127  (not concave)
Iteration 23:  log likelihood =  -1297.235  (not concave)
Iteration 24:  log likelihood = -1296.8556  (not concave)
Iteration 25:  log likelihood = -1296.4713  (not concave)
Iteration 26:  log likelihood = -1296.0906  (not concave)
Iteration 27:  log likelihood = -1295.7245  (not concave)
Iteration 28:  log likelihood = -1295.3545  (not concave)
Iteration 29:  log likelihood = -1294.9892  (not concave)
Iteration 30:  log likelihood = -1294.6246  (not concave)
Iteration 31:  log likelihood = -1294.2688  (not concave)
Iteration 32:  log likelihood = -1293.9114  (not concave)
Iteration 33:  log likelihood = -1293.5606  (not concave)
Iteration 34:  log likelihood = -1293.2094  (not concave)
Iteration 35:  log likelihood = -1292.8657  (not concave)
Iteration 36:  log likelihood =  -1292.521  (not concave)
--Break--
r(1);

 Many thanks,

Yours sincerely,

Phuong

*
*   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/