Forecasting Stock Market Using Machine Learning Approach Encoder-Decoder ConvLSTM

The stock market prediction is a hot topic these days, and predicting the price of a stock is both difficult and important due to the numerous variables at play. There were numerous Machine Learning models submitted for Stock Market Prediction, but Hybrid Models were successful in making accurate pr...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2021 International Conference on Frontiers of Information Technology (FIT) s. 43 - 48
Hlavní autoři: Iqbal, Khurum, Hassan, Ali, Hassan, Syed Shah Mir Ul, Iqbal, Shuaib, Aslam, Faheem, Mughal, Khurrum S
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.12.2021
Témata:
Deep learning
Encoder-Decoder ConvLSTM
Mean square error methods
Measurement
Predictive models
Root mean square
Stock Market Prediction
Stock markets
Time Series
Time series analysis
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The stock market prediction is a hot topic these days, and predicting the price of a stock is both difficult and important due to the numerous variables at play. There were numerous Machine Learning models submitted for Stock Market Prediction, but Hybrid Models were successful in making accurate predictions. The goal of this study is to create a hybrid Deep Learning model (Encoder-Decoder ConvLSTM) to anticipate stock market prices. We employed historical stock price data S and P500 (Daily Prices) from Yahoo's financial website and six months dataset of State Bank of Pakistan (Hourly values). Different prediction models have been tested for the S and P500 dataset which is publicly available and after finding out that the proposed model performed well it has been applied to the SBP dataset as well. The effectiveness of the proposed model has been calculated based on the following performance metrics, root means square error (RMSE), mean absolute error (MAE), mean square error (MSE), and mean absolute percentage error (MAPE). When compared to other comparable studies, the experimental findings indicate that the proposed model has the best performance metrics values. As a result, we can infer that our model is appropriate for accurate stock market time series prediction.
DOI:10.1109/FIT53504.2021.00018