Verify that there was error in the readme file for 2018 models (but models were correctly done)

[andrew-edwards]

Just want to document this. For the 2018 models, the readme file

https://docs.google.com/spreadsheets/d/1OfcrloTP-ctETqLOojSCRNc0qUUMQ3W5k3qfxf5XtNY/edit#gid=0

says, for sensitivity run

2018.40.15_compWeight_HarmonicMean

Change the weighting of composition data to the McAllister-Ianelli (harmonic mean) approach

then in the comments:

using model '2018.40': change control file to add 2 parameter lines after selectivity pars but before 
"timevary selex parameters" and comment out variance adjustments factors at bottom of file. 
In Data file, change "CompError" column in age comp specifications from 0 0 to 1 1 
and "ParmSelect" from 0 0 to 1 2.

From comparing 2018.40_base_model/2018hake_data.ss with 2018.40.15_compWeight_HarmonicMean/2018hake_data.ss it looks like the final bit of instructions in the readme should be the other way round, i.e.:

In Data file, change "CompError" column in age comp specifications from 1 1 to 0 0 
and "ParmSelect" from 1 2 to 0 0.

I just wanted to document this and get it verified at some point. Though writing it out here has helped me clarify it, and I'm pretty sure what I've said is right. (I was double checking our 2019 runs and thought we had the wrong settings for all of them, based on the instructions).

If good we should update the 2019 readme file.

[andrew-edwards]

Looks like the instructions are just from going from the sens case to the base model (I think because in earlier years the base model had the reweighting used in the sens case).

[andrew-edwards]

Be good if @aaronmberger or @taylori can just confirm when they get back. But, yes, close for now...