Token-Based Adaptive Time-Series Prediction by Ensembling Linear and Non-linear Estimators: A Machine Learning Approach for Predictive Analytics on big Stock Data
Bibliographic record
Abstract
With technological advancements, big data can be easily generated and collected in many applications. Embedded in these big data are useful information and knowledge that can be discovered by machine learning and data mining models, techniques or algorithms. A rich source of big data is stock exchange. The ability to effectively predict future stock prices improves the economic growth and development of a country. Traditional linear approaches for prediction (e.g., Kalman filters) may not be practical in handling big data like stock prices due to highly nonlinear and chaotic nature. This lead to the exploitation of various nonlinear estimators such as the extended Kalman filters, expert systems, and various neural network architectures. Moreover, to lessen the potential shortcomings of individual algorithms, ensemble approaches have been created by averaging values across different algorithms. Existing ensemble techniques mostly basket-together a collection of sample-based algorithms that are catered to nonlinear functions. To the best of our knowledge, traditional linear estimators have not yet been incorporated into such an ensemble. Hence, in this paper, we propose a machine learning (specifically, token-based ensemble) algorithm that utilizes both linear and nonlinear estimators to predict big financial time-series data. Our ensemble consists of a traditional Kalman filter, long short-term memory (LSTM) network, and the traditional linear regression model. We also explore the adaptive properties in short-term high-risk trading in the presence of noisy data like stock prices and demonstrate the performance of our ensemble.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".