Optimizing Sequential Predictions for Order Execution: a Decision Focused Learning Approach
Bibliographic record
Abstract
We examine a decision-focused learning (DFL) approach to the problem of optimal liquidation under stochastic liquidity, aiming to optimally integrate time-series forecasting (TSF) models into adaptive liquidation strategies. We consider “Predict-then-Optimize ” style of liquidation strategies that maintain predictions on the future liquidity evolution of the market using TSF models and then liquidate the asset according to the schedule that is repeatedly re-optimized to the prediction. However, when the TSF models are trained to minimize the prediction error, the strategies that integrate these models may fall short in achieving the optimal performance, because the decision maker’s objective is not perfectly aligned with the prediction module’s objective. As a simple remedy, we propose training the TSF models through the DFL approach so that the predictions are adjusted in a way that helps reduce the actual transaction cost. Our suggested framework is agnostic to the TSF model, allowing us to employ (any) modern TSF techniques effortlessly. The numerical experiments on US major stocks show that our suggestion can reduce the suboptimality of naive strategies significantly in one-day liquidation tasks. We further characterize the behaviors of these improved strategies and show that in the synthetic environments the optimal strategy exhibits the similar behaviors.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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".