Optimizing LSTM Based Network For Forecasting Stock Market
Bibliographic record
Abstract
In this modern era, the financial market, more specifically, the stock markets all over the world, deal with an enormous amount of real-time data that facilitates the data analytics and prediction in the field of finance. The main objective of this paper is to propose a novel model of neural network based on Long-Short Term Memory (LSTM) and utilizing one of the most powerful evolutionary algorithms, namely the Differential Evolution (DE), to forecast the next day's stock price of a company. This study focuses on optimizing the ten network hyperparameters related to the detection of temporal patterns of a given dataset, namely, the size of the time window, batch size, the number of LSTM units in hidden layers, the number of hidden layers (LSTM and dense), dropout coefficient for each layer, and the network training optimization algorithm. To the best of our knowledge, this is the first time that all this set of parameters have been optimized simultaneously. Then, the LSTM has been optimized by DE to gain the lower root mean squared error (RMSE) for prediction. The proposed model achieved 8.092 RMSE as its objective value, which is better in comparison with the best statistical forecasting models such as NAIVE, ETS, and SARIMA, which are the-state-of-the-art methods in this filed. Moreover, for shortening the training time as the main source of computational expensiveness, the proposed method works with a lower number of epochs. By this way, DE tries to find a shallower and faster network even with higher accuracy, which is a remarkable approach.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| 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".