st: MLE: Setting Initial Values

["Sabah Abdullah" <sabouha2@gmail.com>]

Dear Stata Listserve:

Presently I am working on MLE and was able to maximize some of my data
but one of them refused to maximize the error message was could not
find feasible values
r(491); (see below for more details) . Hence, I tried to set initial
values but all I am receiving are error messages. Please see below for
my commands and results.
Note I do not have any missing values.  I will be grateful for any
suggestions or comments.

Regards,

Sabah
-- 

COMMANDS:
 capture program drop double_cv_4
 program double_cv_4
 	version 9.2
 	args lnf xb bid
 	qui replace `lnf' = ln(invlogit($ML_y6*`bid'+`xb')) if $ML_y1 == 1
 	qui replace `lnf' = ln(invlogit(-($ML_y7*`bid'+`xb'))) if $ML_y2 == 1
	qui replace `lnf' = ln(invlogit(-($ML_y6*`bid'+`xb')) - ///
 	invlogit(-($ML_y5*`bid'+`xb'))) if $ML_y3 == 1
 	qui replace `lnf' = ln(invlogit(-($ML_y5*`bid'+`xb')) - ///
	invlogit(-($ML_y7*`bid'+`xb'))) if $ML_y4 == 1
 end


ml model lf double_cv_4 (xb: Y_Y_C N_N Y_N N_Y = rev_incm) (bid:
bid_intl bid_low bid_high = )
ml maximize, difficult
ml graph


OUTPUT:
initial:       log likelihood =     -<inf>  (could not be evaluated)

could not find feasible values
r(491);



What I decided to do following recommendation in MLE book by Gould et
al, 3rd edition---loading initial values from vectors:

COMMANDS
. regress Y_Y_C rev_incm sexresp_male

      Source |       SS       df       MS              Number of obs =     200
-------------+------------------------------           F(  2,   197) =    2.52
       Model |  .527840655     2  .263920328           Prob > F      =  0.0827
    Residual |  20.5921593   197  .104528728           R-squared     =  0.0250
-------------+------------------------------           Adj R-squared =  0.0151
       Total |       21.12   199  .106130653           Root MSE      =  .32331

------------------------------------------------------------------------------
       Y_Y_C |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    rev_incm |   2.40e-06   2.92e-06     0.82   0.411    -3.35e-06    8.15e-06
sexresp_male |   .0917239   .0510439     1.80   0.074    -.0089387    .1923866
       _cons |   .0652509   .0386968     1.69   0.093    -.0110622    .1415639
------------------------------------------------------------------------------

. matrix b0 = e(b)

. ml model lf double_cv_4 (xb: Y_Y_C N_N Y_N N_Y = rev_incm
sexresp_male) (bid: bid_intl bid_low bid_high = )  /sigma1
>  /sigma2

. ml init b0

. ml maximize

initial:       log likelihood =     -<inf>  (could not be evaluated)

could not find feasible values
r(491);

. ml check

Test 1:  Calling double_cv_4 to check if it computes log likelihood and
         does not alter coefficient vector...
         Passed.

Test 2:  Calling double_cv_4 again to check if the same log likelihood value
         is returned...
         Passed.

------------------------------------------------------------------------------
The initial values are not feasible.  This may be because the initial values
have been chosen poorly or because there is an error in double_cv_4 and it
always returns missing no matter what the parameter values.

Stata is going to use ml search to find a feasible set of initial values.
If double_cv_4 is broken, this will not work and you will have to press Break
to make ml search stop.

Searching...
initial:       log likelihood =     -<inf>  (could not be evaluated)
searching for feasible values
........................................................................................
> ....................................................................................................................
> ....................................................................................................................
> ....................................................................................................................
> ....................................................................................................................
> ....................................................................................................................
> ....................................................................................................................
> ....................................................................................................................
> ........................................................................................................

could not find feasible values
r(491);


. ml query

Method:           lf
Program:          double_cv_4
Dep. variables:   Y_Y_C N_N Y_N N_Y bid_intl bid_low bid_high
4 equations:
        xb:       rev_incm sexresp_male
       /bid
       /sigma1
       /sigma2
Search bounds:
        xb:            -inf       +inf
       /bid            -inf       +inf
       /sigma1         -inf       +inf
       /sigma2         -inf       +inf
Current (initial) values:
   xb:_cons           35.081399
   bid:_cons          58.844529
   sigma1:_cons       30.126179
   sigma2:_cons       73.279225
    remaining values are zero

.
end of do-file




***********************************************
Ms. Sabah Abdullah
PhD Student
University of Bath
Department of Economics & Int'l Dev.
BATH
BA2 7AY
UNITED KINGDOM
*
*   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/