Performance evaluation metric for statistical learning trading strategies
Bibliographic record
Abstract
We analyze how the sentiment of financial news can be used to predict stock returns and build profitable trading strategies. Combining the textual analysis of financial news headlines and statistical methods, we build multi-class classification models to predict the stock return. The main contribution of this paper is twofold. Firstly, we develop a performance evaluation metric to compare multi-class classification methods, taking into account the precision and accuracy of the models and methods. By maximizing the metric, we find optimal combinations of models and methods and select the best approach for prediction and decision-making. Secondly, this metric enables us to construct profitable option trading strategies, which can also be used as an assessment tool to analyze models' prediction power. We apply our methodology to historical data from Apple stock and financial news headlines from Reuters from January 1, 2012 to May 31, 2019. During validation (May 31, 2018, to May 31, 2019), our models consistently outperformed the market, with two-class one-stage models yielding returns between 30% and 45%, compared to the S & P500 index's 1.73% return over the same period.
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 | 0.028 | 0.068 |
| Meta-epidemiology (narrow) | 0.003 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.007 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.002 |
| 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".