Practical Applications of Weak Supervision and Black–Litterman for Automated ESG Portfolio Construction
Bibliographic record
Abstract
In Weak Supervision and Black–Litterman for Automated ESG Portfolio Construction, published in the Summer 2021 issue of The Journal of Financial Data Science, Alik Sokolov of SR.ai, and Kyle Caverly, Jonathan Mostovoy, Talal Fahoum, and Luis Seco of the University of Toronto’s RiskLab demonstrate the use of machine learning signals for ESG risk in portfolio optimization. The signals are created by a state-of-the-art natural language processing model applied to news articles from The New York Times. The authors demonstrate that the system achieves high accuracy across the ESG categories. They use the Black–Litterman model to combine the signals with a market-weight portfolio to find optimal portfolio weights that reflect the ESG risks. The resulting portfolio outperforms an otherwise equal non-ESG portfolio on a risk-adjusted basis. The approach is promising in that it avoids self-reported biases in the ESG data. Moreover, the results do not imply that risk-adjusted returns must be sacrificed in order to achieve ESG objectives. This area of research is particularly important as interest in ESG investing continues to grow.
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 | 0.008 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 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".