Interpreting financial time series with SHAP values
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
We apply SHAP values to explain how non-linear models predict commentaries on financial time series data. We show how SHAP values are used to assess the usefulness of additional datasets and how they significantly improve the accuracy of tested models. Our industrial partner uses non-linear models to predict commentaries by learning from financial experts reports. Even though a good accuracy has been reached, management wants to demystify the prediction process and needs to demonstrate whether a new and hardly accessible dataset can be useful in prediction. We create an explanation model based on SHAP values to reveal the predominant features and to demonstrate the contribution of the new dataset. This explanation model is also applied to reveal what specific features trigger each class of commentary. We show that new dataset does not improve the learning and that financial experts often rely on specific months to write their commentaries. We also show how SHAP values can be useful in improving the prediction accuracy as they naturally cluster datapoints according to feature importance.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it