Machine learning techniques for cross-sectional equity returns’ prediction
Bibliographic record
Abstract
Abstract We compare the performance of the linear regression model, which is the current standard in science and practice for cross-sectional stock return forecasting, with that of machine learning methods, i.e., penalized linear models, support vector regression, random forests, gradient boosted trees and neural networks. Our analysis is based on monthly data on nearly 12,000 individual stocks from 16 European economies over almost 30 years from 1990 to 2019. We find that the prediction of stock returns can be decisively improved through machine learning methods. The outperformance of individual (combined) machine learning models over the benchmark model is approximately 0.6% (0.7%) per month for the full cross-section of stocks. Furthermore, we find no model breakdowns, which suggests that investors do not incur additional risk from using machine learning methods compared to the traditional benchmark approach. Additionally, the superior performance of machine learning models is not due to substantially higher portfolio turnover. Further analyses suggest that machine learning models generate their added value particularly in bear markets when the average investor tends to lose money. Our results indicate that future research and practice should make more intensive use of machine learning techniques with respect to stock return prediction.
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 | 0.006 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".