put in a parameter into TMB framework, maybe something like local adjustment factor and run model with it, get a mock function for likelihood calculation from TMB and see if the parameter actually has an influence on run model, this would be the end of the proof of concept stage!
put in a parameter into TMB framework, maybe something like local adjustment factor and run model with it, get a mock function for likelihood calculation from TMB and see if the parameter actually has an influence on run model, this would be the end of the proof of concept stage!