Paper: arXiv 2507.01964
Authors: Adebola K. Ojo, Ifechukwude Jude Okafor
Abstract
Investors and stock market analysts face major challenges in predicting stock returns and making wise investment decisions. The predictability of equity stock returns can boost investor confidence, but it remains a difficult task. To address this issue, a study was conducted using a Long Short-term Memory (LSTM) model to predict future stock market movements. The study used a historical dataset from the Nigerian Stock Exchange (NSE), which was cleaned and normalized to design the LSTM model. The model was evaluated using performance metrics and compared with other deep learning models like Artificial and Convolutional Neural Networks (CNN). The experimental results showed that the LSTM model can predict future stock market prices and returns with over 90% accuracy when trained with a reliable dataset. The study concludes that LSTM models can be useful in predicting financial time-series-related problems if well-trained. Future studies should explore combining LSTM models with other deep learning techniques like CNN to create hybrid models that mitigate the risks associated with relying on a single model for future equity stock predictions.
Complexity vs Empirical Score
- Math Complexity: 4.0/10
- Empirical Rigor: 3.5/10
- Quadrant: Philosophers — conceptual discussion, lower math and empirical scores
Why this score: The paper uses standard LSTM architecture without advanced mathematical derivations, fitting a low complexity score; while it reports metrics like accuracy, the lack of specific backtesting details, cross-validation, or live performance data results in low empirical rigor.
Research Flowchart
flowchart TD A["Research Goal:<br>Predict Nigerian Equity Stock Returns"] --> B["Data Source:<br>Historical NSE Dataset"] B --> C["Preprocessing:<br>Data Cleaning & Normalization"] C --> D["Modeling Technique:<br>LSTM (Long Short-Term Memory)"] D --> E["Evaluation & Comparison:<br>Metrics vs. CNN & ANN Models"] E --> F["Outcome:<br>>90% Prediction Accuracy"] F --> G["Conclusion:<br>LSTM is effective for financial time-series forecasting"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes using a Long Short‑Term Memory (LSTM) neural network to forecast Nigerian equity stock returns.
Source sentence
a study was conducted using a Long Short-term Memory (LSTM) model to predict future stock market movements.
Assumes 6
- Only equities with over 20 years of consistent historical data (2001‑2022) are used. data
Source sentence
Only equities with over 20 years of consistent historical dataset (2001-2022) were chosen for the tests in this study
- The dataset is cleaned of noise, missing data and inconsistencies before modeling. data
Source sentence
The dataset was cleaned up by eliminating noise, missing data, and inconsistencies.
- Features are scaled to the [0,1] range using MinMaxScaler. model
Source sentence
This entails producing a scalar object (with feature range of 0, 1 values) using the MinMaxScaler tool.
- 80% of the data is used for training and 20% for validation/testing. execution
Source sentence
... the trained model was test‑proved with 20% of the dataset reserved for validation
Show 2 more
- The LSTM architecture uses 40 units, input shape (55,1) and dropout 0.2. model
Source sentence
The LSTM (Deep learning) algorithm was applied with a unit value of 40, a total input shape of (55,1), and a Dropout value of 0.2.
- Historical price patterns are assumed to be predictive of future returns. market
Source sentence
The experimental results showed that the LSTM model can predict future stock market prices and returns with over 90% accuracy
Data
Nigerian Stock Exchange · equities · daily
- Nigerian Stock Exchange equity dataset market-prices public
Source sentence
Dataset was obtained from Nigerian Stock Exchange website which is made up of equity‑stocks that represent each Nigerian economic sector
Costs
Not applicable
Finds 4
- The LSTM model achieved over 90% prediction accuracy on the test set. predictive-accuracy
Source sentence
The experimental results showed that the LSTM model can predict future stock market prices and returns with over 90% accuracy when trained with a reliable dataset.
- GTBank stock predictions yielded an accuracy of 0.98533, MAE 0.5672, MSE 0.7624, RMSE 0.87318. performance
Source sentence
GTBank 0.98533 0.5672 0.7624 0.87318
- Across selected equities, LSTM outperformed ANN and CNN in accuracy and error metrics. performance
Source sentence
The graph shows pictorially that the LSTM model built did extremely well against some other neural network models such convolutional neural network (CNN), and Artificial Neural Networks (ANN) respectively.
- The LSTM model’s MAE, MSE and RMSE values were low, indicating good prediction performance. statistical
Source sentence
while MAE of value 0.567, MSE value of 0.762 and RMSE of 0.87 respectively shows the loss errors are low, thus a good prediction performance by the LSTM model.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 42 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.