Optimization terminated: relative infinity-norm of gradient less than options.TolFun.

Computing finite-difference Hessian using user-supplied objective function.
ans =
         est.    std.err.
1     0.5596    0.05403  
2     -0.399    0.05738  
3      0.363    0.02664  
4    -0.3399    0.01532  
ans =
LL at convergence: 
-2207              
??? Undefined function or method 'GMNL_2009a' for input arguments of type 'struct'.

Error in ==> <a href="error:c:\gmnl2009b\gmnl_setup.m,60,1">gmnl_setup at 60</a>
    Results_clogit_t = GMNL_2009a(YVEC,XMAT,TIMES,EstimOpt,ModelOpt,[]);

gmnl_setup
Optimization terminated: relative infinity-norm of gradient less than options.TolFun.

Computing finite-difference Hessian using user-supplied objective function.
ans =
         est.    std.err.
1     0.5596    0.05403  
2     -0.399    0.05738  
3      0.363    0.02664  
4    -0.3399    0.01532  
ans =
LL at convergence: 
-2207              
estimating scale heterogeneity model...
No. of draws:500
draw for random scale is truncated normal.
Initial values:[0.56 -0.4 0.36 -0.34 0.1]
no constraints on parameters
All parameters are scaled.
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:8.6553e-005
Function value:-2022.394
Function is concave at stopping point.
Elapsed time is 105.195959 seconds.
ans =
         est.    std.err.
1     0.5889     0.1252  
2    -0.5272     0.1102  
3     0.6675     0.1198  
4    -0.6939     0.1243  
5      1.346     0.1185  
estimating mixed logit model...
with uncorrelated errors...
No. of draws:500
Initial values:[0.56 -0.4 0.36 -0.34 0 0 0 0]
no constraints on parameters
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:4.2104e-005
Function value:-1741.5682
Function is concave at stopping point.
Elapsed time is 39.100067 seconds.
ans =
         est.    std.err.
1     0.9041     0.1048  
2    -0.6556     0.1048  
3     0.6952    0.09035  
4    -0.6732    0.06496  
5      0.792     0.1084  
6    -0.7193     0.1075  
7      0.959    0.09054  
8     0.8037    0.06464  
estimating mixed logit model...
with correlated errors...
No. of draws:500
Initial values:[0.9 -0.66 0.7 -0.67 0.79 -0.72 0.96 0.8 0.1 0.1 0.1 0.1 0.1 0.1]
no constraints on parameters
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:9.548e-005
Function value:-1728.2554
Function is concave at stopping point.
Elapsed time is 193.339969 seconds.
ans =
         est.    std.err.
 1     0.9277      0.113 
 2    -0.6757     0.1098 
 3     0.8057    0.09246 
 4    -0.7372    0.07456 
 5     0.8184     0.1108 
 6    -0.7232     0.1147 
 7     0.8995     0.1089 
 8     0.8228    0.07485 
 9    -0.1426     0.1933 
10     0.2262     0.1352 
11    -0.2641    0.09293 
12     0.3593     0.1115 
13     0.1312     0.1098 
14    0.09562    0.06768 
estimating G-MNL model...
with uncorrelated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
Initial values:[0.9 -0.66 0.7 -0.67 0.79 -0.72 0.96 0.8 0.1 -1]
no constraints on parameters
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:3.4858e-005
Function value:-1731.3856
Function is concave at stopping point.
Elapsed time is 39.455793 seconds.
ans =
         est.    std.err.
 1      1.085     0.1519 
 2    -0.8876     0.1336 
 3     0.7949     0.1093 
 4     -0.783    0.09662 
 5     0.8116     0.1559 
 6    -0.7448     0.1453 
 7      1.125     0.1437 
 8      1.047     0.1089 
 9     0.6721     0.1344 
10     -1.844      1.191 
estimating G-MNL model...
with uncorrelated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
Initial values:[0.56 -0.4 0.36 -0.34 0.1 0.1 0.1 0.1 0.5 0]
Parameters with zeros are constrained to their initial values[1 1 1 1 1 1 1 1 1 0]
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:2.4595e-005
Function value:-1725.4718
Function is concave at stopping point.
Elapsed time is 45.865021 seconds.
ans =
         est.    std.err.
1      1.179     0.1467  
2    -0.8353     0.1328  
3      1.039     0.1133  
4    -0.9061    0.09412  
5     0.7655      0.131  
6     0.7857     0.1313  
7     0.9292     0.1125  
8     0.8373     0.0786  
9     0.8294     0.1246  
estimating G-MNL model...
with uncorrelated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:8.04e-005
Function value:-1725.4705
Function is concave at stopping point.
Elapsed time is 14.970031 seconds.
ans =
         est.    std.err.
 1      1.177      0.153 
 2    -0.8339     0.1362 
 3      1.037     0.1229 
 4     -0.903     0.1164 
 5     0.7646     0.1315 
 6     0.7851     0.1372 
 7     0.9278     0.1234 
 8     0.8361    0.09258 
 9      0.828     0.1322 
10    0.01799     0.5198 
estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:9.8073e-005
Function value:-1717.6975
Function is concave at stopping point.
Elapsed time is 169.948481 seconds.
ans =
         est.    std.err.
 1      1.195     0.1935 
 2    -0.9687     0.1834 
 3     0.9354     0.1497 
 4    -0.9032      0.128 
 5     0.8399     0.1709 
 6    -0.7975     0.1455 
 7      1.078     0.1567 
 8     0.7869      0.115 
 9    0.07393     0.2294 
10     0.1428     0.1618 
11    -0.2376     0.1165 
12     0.2241     0.1426 
13     0.2698     0.1079 
14     0.3581     0.1057 
15      1.005     0.1764 
16     0.1968     0.5574 
estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
Initial values:[0.56 -0.4 0.36 -0.34 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.5 0]
Parameters with zeros are constrained to their initial values[1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0]
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:8.7232e-005
Function value:-1722.1845
Function is concave at stopping point.
Elapsed time is 262.631634 seconds.
ans =
         est.    std.err.
 1      1.253      0.163 
 2    -0.8974     0.1476 
 3      1.052     0.1241 
 4    -0.8165    0.08864 
 5      0.817     0.1512 
 6     0.7465      0.146 
 7      0.977     0.1221 
 8     0.7757    0.08683 
 9     0.1683      0.202 
10     0.2101     0.1564 
11    0.07073    0.09319 
12    0.08102     0.1287 
13   -0.08445     0.1097 
14     0.3405     0.1013 
15     0.9432     0.1365 
estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:7.9761e-005
Function value:-1720.4336
Function is concave at stopping point.
Elapsed time is 511.362894 seconds.
ans =
         est.    std.err.
 1      1.364     0.2481 
 2     -1.014     0.1953 
 3      1.244     0.1946 
 4    -0.9184      0.142 
 5     0.7703     0.1951 
 6     0.7264     0.1597 
 7      0.888     0.1544 
 8     0.8152     0.1167 
 9     0.3479     0.1997 
10     0.4324     0.1406 
11     0.3146     0.1426 
12   -0.07033     0.1766 
13   -0.05507     0.1246 
14     0.4007      0.103 
15      1.206      0.164 
16     0.3264     0.4657 
estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:9.5729e-005
Function value:-1718.7254
Function is concave at stopping point.
Elapsed time is 241.782434 seconds.
ans =
         est.    std.err.
 1      1.114      0.149 
 2    -0.8677     0.1481 
 3       0.92     0.1346 
 4    -0.8732     0.1086 
 5     0.6878     0.1496 
 6      0.688     0.1489 
 7      1.043      0.152 
 8     0.6572    0.08812 
 9     0.1587     0.1953 
10    -0.2623     0.1537 
11    -0.2252    0.09419 
12     0.3281     0.1505 
13    -0.2571     0.1202 
14     0.3698     0.1079 
15     0.9507     0.1516 
16     0.2306     0.4913 

<a href = "matlab: helpview([docroot '/toolbox/optim/msg_csh/optim_msg_csh.map'],'local_min_unconstrained','CSHelpWindow');">Local minimum found</a>.

Optimization completed because the <a href = "matlab: helpview([docroot '/toolbox/optim/msg_csh/optim_msg_csh.map'],'grad_size','CSHelpWindow');">size of the gradient</a> is less than
the default value of the <a href = "matlab: helpview([docroot '/toolbox/optim/msg_csh/optim_msg_csh.map'],'function_tolerance','CSHelpWindow');">function tolerance</a>.

<<a href = "matlab: createExitMsg('fminusub',1.000000e+00,true,true,'fminunc',3.628822e-07,'default',1.000000e-06,0.000000e+00,'',0.000000e+00,0.000000e+00,'',0.000000e+00);">stopping criteria details</a>>


Computing finite-difference Hessian using user-supplied objective function.

ans =

         est.    std.err.
1     0.5596    0.05403  
2     -0.399    0.05738  
3      0.363    0.02664  
4    -0.3399    0.01532  


ans =

LL at convergence: 
-2207              

estimating scale heterogeneity model...
No. of draws:500
draw for random scale is truncated normal.
Initial values:[0.56 -0.4 0.36 -0.34 0.1]
no constraints on parameters
All parameters are scaled.
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:8.6553e-05
Function value:-2022.394
Function is concave at stopping point.
Elapsed time is 45.862565 seconds.

ans =

         est.    std.err.
1     0.5889     0.1252  
2    -0.5272     0.1102  
3     0.6675     0.1198  
4    -0.6939     0.1243  
5      1.346     0.1185  

estimating mixed logit model...
with uncorrelated errors...
No. of draws:500
Initial values:[0.56 -0.4 0.36 -0.34 0 0 0 0]
no constraints on parameters
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:4.2104e-05
Function value:-1741.5682
Function is concave at stopping point.
Elapsed time is 17.400102 seconds.

ans =

         est.    std.err.
1     0.9041     0.1048  
2    -0.6556     0.1048  
3     0.6952    0.09035  
4    -0.6732    0.06496  
5      0.792     0.1084  
6    -0.7193     0.1075  
7      0.959    0.09054  
8     0.8037    0.06464  

estimating mixed logit model...
with correlated errors...
No. of draws:500
Initial values:[0.9 -0.66 0.7 -0.67 0.79 -0.72 0.96 0.8 0.1 0.1 0.1 0.1 0.1 0.1]
no constraints on parameters
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:9.548e-05
Function value:-1728.2554
Function is concave at stopping point.
Elapsed time is 86.893313 seconds.

ans =

         est.    std.err.
 1     0.9277      0.113 
 2    -0.6757     0.1098 
 3     0.8057    0.09246 
 4    -0.7372    0.07456 
 5     0.8184     0.1108 
 6    -0.7232     0.1147 
 7     0.8995     0.1089 
 8     0.8228    0.07485 
 9    -0.1426     0.1933 
10     0.2262     0.1352 
11    -0.2641    0.09293 
12     0.3593     0.1115 
13     0.1312     0.1098 
14    0.09562    0.06768 

estimating G-MNL model...
with uncorrelated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
Initial values:[0.9 -0.66 0.7 -0.67 0.79 -0.72 0.96 0.8 0.1 -1]
no constraints on parameters
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:3.4858e-05
Function value:-1731.3856
Function is concave at stopping point.
Elapsed time is 17.355416 seconds.

ans =

         est.    std.err.
 1      1.085     0.1519 
 2    -0.8876     0.1336 
 3     0.7949     0.1093 
 4     -0.783    0.09662 
 5     0.8116     0.1559 
 6    -0.7448     0.1453 
 7      1.125     0.1437 
 8      1.047     0.1089 
 9     0.6721     0.1344 
10     -1.844      1.191 

estimating G-MNL model...
with uncorrelated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
Initial values:[0.56 -0.4 0.36 -0.34 0.1 0.1 0.1 0.1 0.5 0]
Parameters with zeros are constrained to their initial values[1 1 1 1 1 1 1 1 1 0]
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:2.4594e-05
Function value:-1725.4718
Function is concave at stopping point.
Elapsed time is 20.101292 seconds.

ans =

         est.    std.err.
1      1.179     0.1467  
2    -0.8353     0.1328  
3      1.039     0.1133  
4    -0.9061    0.09412  
5     0.7655      0.131  
6     0.7857     0.1313  
7     0.9292     0.1125  
8     0.8373     0.0786  
9     0.8294     0.1246  

estimating G-MNL model...
with uncorrelated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:8.04e-05
Function value:-1725.4705
Function is concave at stopping point.
Elapsed time is 6.293801 seconds.

ans =

         est.    std.err.
 1      1.177      0.153 
 2    -0.8339     0.1362 
 3      1.037     0.1229 
 4     -0.903     0.1164 
 5     0.7646     0.1315 
 6     0.7851     0.1372 
 7     0.9278     0.1234 
 8     0.8361    0.09258 
 9      0.828     0.1322 
10    0.01799     0.5198 

estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:9.8073e-05
Function value:-1717.6975
Function is concave at stopping point.
Elapsed time is 75.410397 seconds.

ans =

         est.    std.err.
 1      1.195     0.1935 
 2    -0.9687     0.1834 
 3     0.9354     0.1497 
 4    -0.9032      0.128 
 5     0.8399     0.1709 
 6    -0.7975     0.1455 
 7      1.078     0.1567 
 8     0.7869      0.115 
 9    0.07393     0.2294 
10     0.1428     0.1618 
11    -0.2376     0.1165 
12     0.2241     0.1426 
13     0.2698     0.1079 
14     0.3581     0.1057 
15      1.005     0.1764 
16     0.1968     0.5574 

estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
Initial values:[0.56 -0.4 0.36 -0.34 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.5 0]
Parameters with zeros are constrained to their initial values[1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0]
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:8.7228e-05
Function value:-1722.1845
Function is concave at stopping point.
Elapsed time is 116.671655 seconds.

ans =

         est.    std.err.
 1      1.253      0.163 
 2    -0.8974     0.1476 
 3      1.052     0.1241 
 4    -0.8165    0.08864 
 5      0.817     0.1512 
 6     0.7465      0.146 
 7      0.977     0.1221 
 8     0.7757    0.08683 
 9     0.1683      0.202 
10     0.2101     0.1564 
11    0.07073    0.09319 
12    0.08102     0.1287 
13   -0.08445     0.1097 
14     0.3405     0.1013 
15     0.9432     0.1365 

estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:7.9761e-05
Function value:-1720.4336
Function is concave at stopping point.
Elapsed time is 227.300843 seconds.

ans =

         est.    std.err.
 1      1.364     0.2481 
 2     -1.014     0.1953 
 3      1.244     0.1946 
 4    -0.9184      0.142 
 5     0.7703     0.1951 
 6     0.7264     0.1597 
 7      0.888     0.1544 
 8     0.8152     0.1167 
 9     0.3479     0.1997 
10     0.4324     0.1406 
11     0.3146     0.1426 
12   -0.07033     0.1766 
13   -0.05507     0.1246 
14     0.4007      0.103 
15      1.206      0.164 
16     0.3264     0.4657 

estimating G-MNL model...
with correlated errors...
No. of draws:500
draw for random scale is truncated normal. the rest are standard normal.
All parameters are scaled.
setting LBVAR for gamma* to -5
setting UBVAR for gamma* to 5
BHHH with analytical gradients
**=============================================================================================**
Converged with tolerance:9.5729e-05
Function value:-1718.7254
Function is concave at stopping point.
Elapsed time is 110.676985 seconds.

ans =

         est.    std.err.
 1      1.114      0.149 
 2    -0.8677     0.1481 
 3       0.92     0.1346 
 4    -0.8732     0.1086 
 5     0.6878     0.1496 
 6      0.688     0.1489 
 7      1.043      0.152 
 8     0.6572    0.08812 
 9     0.1587     0.1953 
10    -0.2623     0.1537 
11    -0.2252    0.09419 
12     0.3281     0.1505 
13    -0.2571     0.1202 
14     0.3698     0.1079 
15     0.9507     0.1516 
16     0.2306     0.4913 

