5 LIMITATIONS AND FUTURE
OUTLOOKS
The present study, while demonstrating the potential
of LSTM networks in predicting the S&P 500 index
with the augmentation of financial factors, is not
without its limitations. One key limitation lies in the
reliance on a simple LSTM architecture. As the
financial forecasting landscape evolves rapidly,
exploring alternative LSTM variants, such as stacked
or bidirectional LSTMs, or hybrid architectures
combining LSTMs with CNNs or attention
mechanisms, could potentially enhance predictive
capabilities. Additionally, the evaluation framework,
utilizing RMSE, MAE, and classification accuracy,
provides valuable insights but may be further refined
by incorporating metrics like R-squared for
regression or F1-score for imbalanced classification
problems.
Looking ahead, the dynamic nature of financial
markets necessitates mechanisms for model
retraining and adaptation to maintain predictive
accuracy over time. Continuous monitoring of market
dynamics and regular updating of model parameters
are crucial. Moreover, there may be other relevant
variables, such as financial news, economic
indicators, or sentiment data from social media, that
could be incorporated to improve predictive
power. Future research should aim to address these
limitations by exploring alternative architectures,
refining evaluation metrics, incorporating additional
data sources, and implementing mechanisms for
continuous model updating.
6 CONCLUSIONS
This study has shown the potential ability of LSTM
networks to predict the S&P 500 index, particularly
when augmented with financial factors. The findings
underscore the effectiveness of LSTM models in
capturing short-term market fluctuations, evidenced
by their relatively low RMSE and MAE values for 1-
day predictions. However, as the study also
highlights, predicting longer-term trends remains a
challenge, with errors increasing for 5-day and 20-
day horizons, especially for variables sensitive to
market volatility and interest rate changes. Looking
toward the future, it is crucial to acknowledge that the
dynamic nature of financial markets necessitates
ongoing efforts to maintain predictive accuracy. This
includes exploring alternative LSTM variants and
hybrid architectures, refining evaluation metrics,
incorporating additional information sources
encompassing financial updates, economic metrics,
and public opinion reflected on social media
platforms, and implementing mechanisms for
continuous model updating and adaptation. By
addressing these limitations and harnessing the full
potential of LSTM networks, the model can further
enhance the ability to forecast the S&P 500 index
while providing valuable insights for investors,
portfolio managers, policymakers, and so on.
REFERENCES
Atsalakis, G. S., Valavanis, K. P., 2009. Surveying stock
market forecasting techniques–Part II: Soft computing
methods. Expert Systems with applications,36(3),
5932-5941.
Bhandari, H. N., Rimal, B., Pokhrel, N.R., Rimal, R.,
Dahal, K. R., & Khatri, R. K., 2022. Predicting stock
market index using LSTM. Machine Learning with
Applications,9, 100320.
Ding, G., Qin, L., 2020. Study on the prediction of stock
price based on the associated network model of LSTM.
International Journal of Machine Learning and
Cybernetics, 11(6), 1307-1317.
Lee, J., Kang, J., 2020. Effectively training neural networks
for stock index prediction: Predicting the S&P 500
index without using its index data. PloS one, 15(4),
e0230635.
Michańków, J., Sakowski, P., Ślepaczuk, R., 2022. LSTM
in algorithmic investment strategies on BTC and
S&P500 index. Sensors, 22(3), 917.
Mehtab, S., Sen, J., Dutta, A., 2021. Stock price prediction
using machine learning and LSTM-based deep learning
models. In Machine Learning and Metaheuristics
Algorithms, and Applications: Second Symposium,
SoMMA 2020, Chennai, India, October 14–17, 2020,
Revised Selected Papers 2 (pp.88-106). Springer
Singapore.
Pramod, B. S., Pm, M. S., 2020. Stock price prediction
using LSTM.Test Engineering and Management, 83,
5246-5251.
Roszyk, N., Ślepaczuk, R., 2024. The Hybrid Forecast of
S&P 500 Volatility ensembled from VIX, GARCH and
LSTM models.arxiv preprint arxiv:2407.16780.
Vargas, M. R., De Lima, B. S., Evsukoff, A. G., 2017. Deep
learning for stock market prediction from financial
news articles. In 2017 IEEE international conference on
computational intelligence and virtual environments for
measurement systems and applications (CIVEMSA)
(pp. 60-65). IEEE.
Vo, N., Ślepaczuk, R., 2022. Applying hybrid ARIMA-
SGARCH in algorithmic investment strategies on
S&P500 index. Entropy, 24(2), 158.
Wang, C., Chen, Y., Zhang, S., Zhang, Q., 2022. Stock
market index prediction using deep Transformer model.
Expert Systems with Applications, 208, 118128.