- Data and Parameters: The
dataandparlists contain the input data and parameters, respectively. These are passed to theobjective_function. - Objective Function: The
objective_functioncomputes the negative log-likelihood (nll) and returns a list of results, including likelihood components, predictions, and residuals. - RTMB Compatibility: The model is written in R but uses RTMB to handle the C++ backend, making it compatible with TMB's optimization and simulation tools.
- Report: The
reportlist contains all the outputs, including likelihood components, predictions, and residuals.
This setup allows you to use the model in R while leveraging the performance of TMB's C++ backend. You can optimize the model using RTMB::MakeADFun and RTMB::nlminb.