An empirical comparison of alternative likelihood formulations for composition data, with application to cobia and Pacific hake
Bibliographic record
Abstract
Fitting composition data within stock assessment models has historically utilized the multinomial likelihood, often with iterative reweighting algorithms to account for overdispersion due to sampling and process error. Recently, the Dirichlet-multinomial has been increasingly incorporated into assessments as a composition likelihood that can be internally weighted using an estimated overdispersion parameter. There exist two popular formulations of the Dirichlet-multinomial. Recent research has also suggested improved performance in assessments using the logistic-normal for composition data, specifically when the composition sample size is large. We evaluated the performance of two Dirichlet-multinomial formulations and the logistic-normal by incorporating them into assessments that differed greatly in sample sizes for composition data: cobia ( Rachycentron canadum) and Pacific hake ( Merluccius productus). We compared the likelihoods against one another using various model diagnostic criteria common in stock assessments. Overall, the linear formulation of the Dirichlet-multinomial outperformed the saturating formulation. At small sample sizes of the cobia assessment, the logistic-normal performed poorly. The comparison was more robust at large sample sizes of the Pacific hake assessment; however on balance, it seems prudent to proceed with the Dirichlet-multinomial.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.067 | 0.199 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.004 | 0.006 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".