Hi Jintao,
Just a few changes to the control file were necessary to get the
model running.
1. I changed AUTOGEN from
0 0 0 0 0 # autogen: 1st element for biology, 2nd for SR, 3rd for Q,
4th reserved, 5th for selex
to
1 0 0 0 0 # autogen: 1st element for biology, 2nd for SR, 3rd for Q,
4th reserved, 5th for selex
(Note: use all 1s to turn off autogen for all parameter sections)
2. The M parameter line needs the link and then parameter number
specified in a 3 digit number, so
0.2 0.4 0.23 0.1 0.8 0 -3 1 0 0 0 0 0 0 # NatM_p_1_Fem_GP_1
became
0.2 0.4 0.23 0.1 0.8 0 -3 201 0 0 0 0 0 0 # NatM_p_1_Fem_GP_1
(the use of a 2 means an additive link, see the 3.30.17 user manual
for other link options.)
This should be enought to get the model running, however there were a
few additional steps to make the model make more sense:
3. Loosen the bounds on the M paramter line to allow higher values of
M than 0.4.
0.2 0.4 0.23 0.1 0.8 0 -3 201 0 0 0 0 0 0 # NatM_p_1_Fem_GP_1
becomes
0.2 2 0.23 0.1 0.8 0 -3 201 0 0 0 0 0 0 # NatM_p_1_Fem_GP_1
4. The short parameter line represents the scalar on the
environmental link when using the additive link (See using
time-varying paramters in the 3.30.17 user manual for the equations):
M for the year = base M in long parameter line (0.23 in this case) +
M in datafile for the year * short parameter line scalar
In order to just add the values in the data file directly to the base
M, I changed:
0.2 0.4 0.23 0.1 0.8 0 -3 #_placeholder when no time-vary parameters
to
0 2 1 0.1 0.8 0 -3 # scalar on the env parameter.
So that 1 is the scalar. This means the final value of M for the year
is just the base paramter + the value for the year in the data file.
To verify the values used in the model, they should be in a table in
the report file labelled : MGparm_By_Year_after_adjustments . There
should also be a plot of M generated using r4ss::SS_plots
Feel free to post again if I have left any questions unanswered, the
setup for using time varying parameters can be complex.
Hope this helps,
Kathryn