User Tools

Site Tools


admb_real:monotonizing_surveys

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
admb_real:monotonizing_surveys [2010/09/21 14:33] lennartfradmb_real:monotonizing_surveys [2010/09/21 14:44] (current) lennartfr
Line 1: Line 1:
 ===== Making survey data monotonic ===== ===== Making survey data monotonic =====
-As one can observe in [[admb_real:vpa_estimates|plots of the survey indices]], these are frequently not monotonic. For this reason we make the survey indices nonicreasing by removing the offending indides, and repeat the experiments in the section about [[admb_real:two_qs_cohorts_per_run|choosing the number of cohorts per run]]. To be more specific, we remove the observations of a cohort which are smaller than subsequent observations.+As one can observe in plots of the survey indices [[admb_real:sdreport_and_vpa|(see top here)]], these are frequently not monotonic. For this reason we make the survey indices nonicreasing by removing the offending indides, and repeat the experiments in the section about [[admb_real:two_qs_cohorts_per_run|choosing the number of cohorts per run]]. To be more specific, we remove the observations of a cohort which are smaller than subsequent observations.
  
 === Summary === === Summary ===
Line 165: Line 165:
  
 ^ Num. cohorts ^ Trajectories ^ Estimates ^ ^ Num. cohorts ^ Trajectories ^ Estimates ^
-| 1 | |<file txt cod.par> </file> +| 1 |{{:admb_real:cod_two_q_mono_vs_vpa_97_97.png?180|}} |<file txt cod.par># Number of parameters = 5  Objective function value = -0.576982  Maximum gradient component = 6.53503e-06 
-| 2 | |<file txt cod.par> </file> +# N0: 
-| 3 | |<file txt cod.par> </file> + 3.18481 
-| 4 | |<file txt cod.par> </file>  |+# q02: 
 + 0.572871 
 +# q3plus: 
 + 0.333259 
 +# logs: 
 +-0.572122768699 
 +# M: 
 +0.389392591051 
 + </file> 
 +| 2 |{{:admb_real:cod_two_q_mono_vs_vpa_97_98.png?180|}} |<file txt cod.par># Number of parameters = 6  Objective function value = -2.56523  Maximum gradient component = 2.18165e-05 
 +# N0: 
 + 28.7168 18.0327 
 +# q02: 
 + 0.0500009 
 +# q3plus: 
 + 0.0543061 
 +# logs: 
 +-0.660327137031 
 +# M: 
 +0.499995000270 
 + </file> 
 +| 3 |{{:admb_real:cod_two_q_mono_vs_vpa_97_99.png?180|}} |<file txt cod.par># Number of parameters = 7  Objective function value = -5.07237  Maximum gradient component = 8.93392e-05 
 +# N0: 
 + 23.7926 14.7673 9.86596 
 +# q02: 
 + 0.0500001 
 +# q3plus: 
 + 0.0544865 
 +# logs: 
 +-0.720537843156 
 +# M: 
 +0.464181122587 
 + </file> 
 +| 4 |{{:admb_real:cod_two_q_mono_vs_vpa_97_00.png?180|}} |<file txt cod.par># Number of parameters = 8  Objective function value = -8.16818  Maximum gradient component = 4.40661e-06 
 +# N0: 
 + 21.6570 13.2731 8.74892 18.7506 
 +# q02: 
 + 0.0517102 
 +# q3plus: 
 + 0.0500001 
 +# logs: 
 +-0.772272656988 
 +# M: 
 +0.435222747275 
 + </file>  |
  
  
admb_real/monotonizing_surveys.1285079581.txt.gz · Last modified: 2010/09/21 14:33 by lennartfr