[Limdep Nlogit List] Grouped data error message
Andreas C Drichoutis
adrihout at aua.gr
Wed Aug 9 17:57:44 EST 2006
Hi all,
I'm estimating a grouped data model and I get the below error message along
with a normal exit from iterations status. Are results trustworthy?
Thanks,
Andreas
--> grouped ; lhs=VNL ; rhs=ONE, TAST, NUT, WWEEKH, STR,
NFLX, PHDM, ISMED, ISELSE, ISFRIEN,ISNO, PLANNER,WEST,EAST, SOUTH,
NORTH, BH,GH, EXER, NKN,NI,EFF2,GEND,AGE, EDUC2,EDUC3,
INC1,INC2,INC3,ALWOFT, MED, OV,OB, HORIZON
; limits=0,2,3,5,8,13,21,36,60 ; KEEP=FITWTP2 $
Hessian is not positive definite at start values.
Error: 803: Hessian is not positive definite at start values.
B0 is too far from solution for Newton method.
Switching to BFGS as a better solution method.
Error: 137: Iterations: function not computable at crnt. trial estimate
Normal exit from iterations. Exit status=0.
+---------------------------------------------+
| Limited Dependent Variable Model - CENSORED |
| Maximum Likelihood Estimates |
| Model estimated: Aug 09, 2006 at 10:50:06AM.|
| Dependent variable VNL |
| Weighting variable None |
| Number of observations 262 |
| Iterations completed 40 |
| Log likelihood function -542.1200 |
| Censoring Thresholds for the 10 cells: |
| Lower Upper Lower Upper |
| 1 ******* .00 2 .00 2.00 |
| 3 2.00 3.00 4 3.00 5.00 |
| 5 5.00 8.00 6 8.00 13.00 |
| 7 13.00 21.00 8 21.00 36.00 |
| 9 36.00 60.00 10 60.00 ******* |
+---------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
Primary Index Equation for Model
Constant -8.482913005 10.233783 -.829 .4072
TAST 5.478935157 4.9333578 1.111 .2667 .97328244
NUT 2.177748079 3.8330831 .568 .5699 .95038168
WWEEKH .1680148164 .57624526E-01 2.916 .0035 17.062977
STR .8091784428 3.3797176 .239 .8108 .72519084E-01
NFLX -7.947525292 2.5603588 -3.104 .0019 .18702290
PHDM 4.183857111 2.5979839 1.610 .1073 .12595420
ISMED .7586809924 2.9977169 .253 .8002 .76335878E-01
ISELSE -3.926185490 4.6233206 -.849 .3958 .34351145E-01
ISFRIEN -.3896667449 2.0310790 -.192 .8479 .20229008
ISNO 2.533267410 2.2699547 1.116 .2644 .17938931
PLANNER -3.313984298 2.6051747 -1.272 .2033 .76717557
WEST -5.941326081 2.5333251 -2.345 .0190 .29770992
EAST 2.170737846 3.1197779 .696 .4866 .10687023
SOUTH -1.999043292 2.6000425 -.769 .4420 .21755725
NORTH -4.495901729 2.6671186 -1.686 .0919 .20992366
BH -.8803782412 2.8059552 -.314 .7537 .11832061
GH -1.881621104 1.8778815 -1.002 .3163 .50381679
EXER -.9005623461 2.0039123 -.449 .6531 .25954198
NKN 1.296717302 .65778032 1.971 .0487 5.5190840
NI -.6860157977 .71955480 -.953 .3404 4.5725191
EFF2 2.095537087 3.1710858 .661 .5087 .69338422
GEND -2.860793117 2.5055022 -1.142 .2535 .32061069
AGE .4139288739E-01 .67178016E-01 .616 .5378 49.572519
EDUC2 1.829472402 2.1265093 .860 .3896 .45419847
EDUC3 -1.524310636 2.8037247 -.544 .5867 .29007634
INC1 4.408267518 3.5997424 1.225 .2207 .22519084
INC2 -1.285378004 3.5167576 -.366 .7147 .33969466
INC3 1.958109458 3.2865041 .596 .5513 .34732824
ALWOFT 3.913428558 1.7985271 2.176 .0296 .34351145
MED 3.683483871 2.6087065 1.412 .1580 .11450382
OV -4.368004327 1.8488591 -2.363 .0182 .41603053
OB -.8965140373 2.2921617 -.391 .6957 .18320611
HORIZON 4.820180890 1.5867367 3.038 .0024 .38167939
Sigma 11.23682290 .64048764 17.544 .0000
(Note: E+nn or E-nn means multiply by 10 to + or -nn power.)
More information about the Limdep
mailing list