A Long Short-Term Memory Network Stock Price Prediction with Leading Indicators
Bibliographic record
Abstract
The accuracy of the prediction of stock price fluctuations is crucial for investors, and it helps investors manage funds better when formulating trading strategies. Using forecasting tools to get a predicted value that is closer to the actual value from a given financial data set has always been a major goal of financial researchers and a problem. In recent years, people have paid particular attention to stocks, and gradually used various tools to predict stock prices. There is more than one factor that affects financial trends, and people need to consider it from all aspects, so research on stock price fluctuations has also become extremely difficult. This paper mainly studies the impact of leading indicators on the stock market. The framework used in this article is proposed based on long short-term memory (LSTM). In this study, leading indicators that affect stock market volatility are added, and the proposed framework is thus named as a stock tending prediction framework based on LSTM with leading indicators (LSTMLI). This study uses stock markets in the United States and Taiwan, respectively, with historical data, futures, and options as data sets to predict stock prices in these two markets. We measure the predictive performance of LSTMLI relative to other neural network models, and the impact of leading indicators on stock prices is studied. Besides, when using LSTMLI to predict the rise and fall of stock prices in the article, the conventional regression method is not used, but the classification method is used, which can give a qualitative output based on the data set. The experimental results show that the LSTMLI model using the classification method can effectively reduce the prediction error. Also, the data set with leading indicators is better than the prediction results of the single historical data using the LSTMLI model.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 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 | 0.000 |
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".