A simulation/estimation procedure to test if a stock assessment model can estimate the catch and proportions-at-age, or catch-at-age under an array of assumed distributions of the observed data
Options for assumed distribution:
- Multivariate Log Normal (Catch-at-age)
- Univariate Log normal (Catch) and Multinomial (Proportions-at-age)
- Univariate Log normal (Catch) and Dirichlet Multinomial (Proportions-at-age)
Wish list for additional distributions: 4. Univariate Log normal (Catch) and Logistic Normal (Proportions-at-age) 5. Univariate Log normal (Catch) and multivariate-Tweedie (Proportions-at-age)
Resources/Inspiration: Thorson, James T., et al. "Model-based estimates of effective sample size in stock assessment models using the Dirichlet-multinomial distribution." Fisheries Research 192 (2017): 84-93.
Fisch, Nicholas, et al. "Assessing likelihoods for fitting composition data within stock assessments, with emphasis on different degrees of process and observation error." Fisheries Research 243 (2021): 106069.
Albertsen, Christoffer Moesgaard, Anders Nielsen, and Uffe Høgsbro Thygesen. "Choosing the observational likelihood in state-space stock assessment models." Canadian Journal of Fisheries and Aquatic Sciences 74.5 (2017): 779-789.