A Novel Parallel Recurrent Fusion Network for Stock Market Forecasting
Bibliographic record
Abstract
Stock market forecasting is inherently difficult due to the volatile and nonlinear nature of financial data. Traditional statistical models, such as ARIMA, often fail to capture long-term dependencies, while deep learning models-particularly Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU), have shown better predictive performance. However, single-model architectures do not usually tap the various benefits provided by different recurrent networks. In this work, we proposed the Parallel Recurrent Fusion Network (PRFN), a novel hybrid model that combines the strengths of RNN, LSTM, and GRU to improve stock price forecasting accuracy without compromising computational efficiency. The model was trained on 16 years (5,844 days) of historical stock data from Reliance Industries Ltd, normalized and segmented with a sliding window (60-day lookback and 7-day forecast horizon). Performance evaluation included Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Coefficient of Determination (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) for direct multi-step forecasting, sequence-to-sequence forecasting, and the proposed hybrid model. The experimental results reveal that the RNN-based direct multi-step model surpassed standalone LSTM and GRU models by achieving R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.939 and RMSE of 35.39, while the proposed PRFN (RNN + GRU) gave us an RMSE of 34.52 and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.942. These findings suggest that the concurrent use of hybrid deep learning models could greatly improve stock price predictability, all while melding together the many strengths of different recurrent architectures.
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 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 | 0.014 | 0.029 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| 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 teacher head, 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".