- Univariate Time Series
Predicting stock price using historical data of a company using Neural Networks for multi-step forecasting of stock price.
This project is; to implement deep learning algorithms two sequential models of recurrent neural networks (RNNs) such as stacked LSTM, Bidirectional LSTM, and NeuralProphet built with PyTorch to predict stock prices using time series forecasting.
- What is Time Series?
- What is LSTM?
- What is Bidirectional LSTM?
- What is NeuralProphet?
- Started With the Stock Data
- Model Implementation Phase
- Models Train & validation Loss
- Conclusion
This project was created for the article: Univariate Time Series With Stacked LSTM, BiLSTM, and NeuralProphet.
Data are obtained from 2010–01–04 to 2021–11–02 (11 years, 9 months, and 29 days) for Apple Inc (AAPL) and exported directly from Yahoo finance. Stock price history will be for the past 11 years (including the Covid-19 period).
- Use Yahoo Finance API to grab stock data.
- More about data descrption from here
- Stacked LSTM
- Bidirectional LSTM
- NeuralProphet
- Colab
- numpy
- pandas
- sklearn
- scipy
- matplotlib
- tensorflow
- keras
- Plotly
- NeuralProphet
- h5py
- Facebook Prophet (Included to the project)
The analysis of using three deep learning algorithms shows that they remain a worthy investment for the future. A qualified investor, on the other hand, should perform both external and internal business research before making an investment decision in order to obtain a full picture of the company's potential value.
Attempts have been made to predict stock prices using time series analysis algorithms, but they are not yet available for betting in the real market. This is just a tutorial and implementation of deep learning models to forecast stock. Therefore, it is not intended to instruct people to buy stock from this repo.
You can check out the full contribution document here
- Fork the Repository here
1- Fork it
2- Create your feature branch (git checkout -b feature/fooBar)
3- Commit your changes (git commit -am 'Add some fooBar')
4- Push to the branch (git push origin feature/fooBar)
5- Create a new Pull Request
Stock Price Prediction Using Deep Learning - feel free to contact!