[ADMB Users] Fwd: lgam in 10.0

Sun May 15 15:51:45 PDT 2011

Thanks Dave, the other word order sort of worked. Will these get standardized in a future release? Maybe I could work on that if I attend the developers meeting.
Now the problem is that parameters are going negative that should be positive and I have bounded. This is really strange. Maybe it's pow again? I didn't end up fixing the gradient problem from the separate post, because this problem was more urgent. I wonder if my newly compiled version is broken.

mollie-brookss-macbook-4:FirFecundity molliebrooks$ admb firdbhfec

*** Parsing: tpl2cpp firdbhfec

*** Compiling: CXXFLAGS="-m64" adcomp firdbhfec
g++ -c -O3 -Wno-deprecated  -DOPT_LIB -D__GNUDOS__ -Dlinux -DUSE_LAPLACE -fpermissive -I. -I/usr/local/admb/include -m64 firdbhfec.cpp

*** Linking: adlink firdbhfec
g++ -O3 -L/usr/local/admb/lib firdbhfec.o -ldf1b2o -ladmod -ladt -lado -ldf1b2o -ladmod -ladt -lado -o firdbhfec

Successful build of executable: firdbhfec
mollie-brookss-macbook-4:FirFecundity molliebrooks$ ./firdbhfec

Initial statistics: 3 variables; iteration 0; function evaluation 0; phase 1
Function value   4.7673600e+05; maximum gradient component mag   2.9086e+37
Var   Value    Gradient   |Var   Value    Gradient   |Var   Value    Gradient   
  1 -0.59033 -2.58340e+22 |  2 -0.59033 -6.15817e+22 |  3 -0.59033  2.90865e+37
  ic > imax  in fminim is answer attained ?
Var   Value    Gradient   |Var   Value    Gradient   |Var   Value    Gradient   
  1 -0.59033 -2.58340e+22 |  2 -0.59033 -6.15817e+22 |  3 -0.59033  2.90865e+37
Function minimizer not making progress ... is minimum attained?
Minimprove criterion =   0.0000e+00

 - final statistics:
3 variables; iteration 1; function evaluation 30
Function value   4.7674e+05; maximum gradient component mag   2.9086e+37
Exit code = 1;  converg criter   1.0000e-04
Var   Value    Gradient   |Var   Value    Gradient   |Var   Value    Gradient   
  1 -0.59033 -2.58340e+22 |  2 -0.59033 -6.15817e+22 |  3 -0.59033  2.90865e+37
Estimating row 1 out of 3 for hessian
Estimating row 2 out of 3 for hessian
Estimating row 3 out of 3 for hessian
tau <=1 in log_negbinomial_density 

DATA_SECTION
	init_int nobs
	init_ivector ncones(1,nobs)
	init_vector DBH(1,nobs)
PARAMETER_SECTION
	init_bounded_number a(0.0,10.0)
	init_bounded_number b(0.0,10.0)
	init_bounded_number k(0.0,10.0)
	vector mu(1,nobs)
	objective_function_value f
PROCEDURE_SECTION
	f=0.0;
	mu=a*pow(DBH,b);
	for(int i=1; i<=nobs; i++)
	{
		f-=log_negbinomial_density(ncones[i], mu[i], k);
	}

# nobs 
 242 

# ncones 
 19 42 40 68 5 0 21 114 37 92 84 102 98 63 9 31 35 216 27 297 36 127 23 46 27 66 11 20 141 3 22 39 96 206 40 231 64 202 54 32 108 142 82 65 153 123 131 43 98 37 34 10 65 35 50 19 73 33 61 9 146 0 44 42 0 61 17 53 27 0 74 36 28 56 46 0 15 26 46 15 105 0 62 24 25 41 138 77 228 28 45 57 109 28 17 91 69 87 10 65 50 27 30 86 119 22 8 54 104 14 16 5 53 40 32 114 39 37 111 226 156 42 86 94 54 1 14 44 108 116 14 73 3 16 87 61 48 0 17 5 88 11 133 121 166 171 63 23 4 51 10 14 78 47 31 42 24 42 55 19 63 127 9 74 120 85 51 19 131 7 23 7 9 23 55 48 13 2 9 3 4 16 1 88 8 27 16 184 14 22 25 52 2 134 81 85 3 56 17 8 10 6 69 58 1 22 3 11 22 2 37 8 15 61 6 18 9 109 54 4 11 30 0 0 3 0 16 22 9 56 17 64 38 59 37 22 41 1 22 16 17 4 

# DBH 
 9.4 10.6 7.7 10.6 8.7 10.1 8.1 11.6 10.1 13.3 10 13.4 9.7 7.4 8.7 8.6 7.9 14.2 9.5 15.9 6 10.6 7.3 10.3 8.4 10.2 13.8 9.4 8.1 9.6 7.3 7.4 10.3 13.4 9.2 13.9 10.9 17.4 10.2 8.2 11.3 16.1 12.3 8.3 12.4 12.5 11.3 7.8 11.6 10 7 5.7 7.7 8.9 8.5 8.5 10.7 10.2 10.8 9 9.4 7.6 10.6 10 8 7.4 9.1 6.7 9.7 6.8 8.6 9.1 6.3 6.7 10.9 9.5 9.9 6.8 9.8 7.7 12.1 8.2 10 9.6 9.2 8.2 11.3 11.6 15.7 9.1 8.9 8.7 11 6.6 7.1 9 12.4 12.1 7.5 9 8 10.9 9.2 10.1 12.1 7 6.8 8.6 11.6 6.6 6.7 6.8 8.5 7.8 7.9 9.8 6.2 6.7 15.4 9.2 12.9 6.7 9.6 8.4 8 8.7 6.7 9.2 9.5 8 5.5 8.5 5.7 5.6 8 6.5 9.6 6.1 7.9 5.9 11 8.2 12.8 12.8 12.5 13.7 11.8 6.3 6.3 8.2 6.2 6.7 9.8 9.4 6.7 6 4.9 9.6 7.5 8.4 7.4 9.9 7.4 9.5 13.9 6.9 9.4 7.4 12.8 5.8 7.2 5.6 6.9 11.3 9.6 6.8 6.9 6.6 4.8 4.4 4.8 8.5 7 8.7 6.6 8.6 5.3 10.4 6.4 5.4 8.2 5.5 6.2 14.7 10.5 14.4 5.8 6.1 6.2 6.2 7.2 6 10.6 8.7 7.5 7.3 5.2 6.9 6.6 6.7 5.2 6.9 7.5 9 5.9 6.5 6.6 9.8 4.7 4.2 4.8 6.7 6.5 6.7 5.9 5.4 6.9 6.5 6 12 7.5 6.4 7.3 7.3 6.4 7 5.9 9.1 6.7 4 6.5 4.7 

On 15 May 2011, at 5:31 PM, dave fournier wrote:

> I think it was my lesdyxia acting up.  log_density_negbinomial vs log_negbinomial_density.
> _______________________________________________
> Users mailing list
> Users at admb-project.org
> http://lists.admb-project.org/mailman/listinfo/users
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.admb-project.org/pipermail/users/attachments/20110515/a53f1f07/attachment.html>