Scott Merryman replied:
> Change the -ml model- statement to:
>
> ml model lf mlbeta_ll (`lhs'=`rhs', `nocons') (`dispersion') `if'
> `in', `robust' `svy'
>
>
> Example:
>
> . sysuse auto, clear
> (1978 Automobile Data)
>
> . qui sum mpg
>
> . gen m2 = mpg/(r(max)+1)
>
> . mlbeta m2 trunk gear, nocons
>
> Number of obs
> = 74
> Wald chi2(2)
> = 44.06
> Log likelihood = 54.205681 Prob > chi2
> = 0.0000
>
> -----------------------------------------------------------------------
> -------
> m2 | Coef. Std. Err. z P>|z| [95% Conf.
> Interval]
> -------------+---------------------------------------------------------
> -------
> eq1 |
> trunk | -.0662481 .0104036 -6.37 0.000 -.0866388 -
> .0458574
> gear_ratio | .3262698 .0493823 6.61
> 0.000 .2294823 .4230573
> -------------+---------------------------------------------------------
> -------
> eq2 |
> _cons | 2.81053 .1598448 17.58 0.000 2.49724
> 3.12382
> -----------------------------------------------------------------------
> -------
>
> Mean squared error:.01030075
That didn't work, either. Same result:
. mlbeta sqgovote conadmin labadmin sqgpchg sqbeldm weakgovt, robust nocons
note: labadmin dropped due to collinearity
initial: log pseudolikelihood = 58.897248
...etc
I also took care to run this solution _without_ the two dummy variables in
my model. Guess what:
. mlbeta sqgovote sqgpchg sqbeldm weakgovt, robust nocons
initial: log pseudolikelihood = 58.897248
[...]
Iteration 5: log pseudolikelihood = 218.01871
Number of obs = 118
Wald chi2(3) = 59.26
Log pseudolikelihood = 218.01871 Prob > chi2 = 0.0000
----------------------------------------------------------------------------
| Robust
sqgovote | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-----------+----------------------------------------------------------------
eq1 |
sqgpchg | .8262949 .1342193 6.16 0.000 .56323 1.08936
sqbeldm | -.9181762 .1760294 -5.22 0.000 -1.263188 -.5731649
weakgovt | .1279025 .1039549 1.23 0.219 -.0758455 .3316504
_cons | -2.52928 .2697912 -9.37 0.000 -3.058061 -2.000499
-----------+----------------------------------------------------------------
eq2 |
_cons | 3.066009 .3706643 8.27 0.000 2.339521 3.792498
----------------------------------------------------------------------------
Mean squared error: .00640215
Sorry, Scott!
CLIVE NICHOLAS |t: 0(044)7903 397793
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
*
* 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/
