This repository is a comprehensive project focused on time series analysis and Support Vector Machine (SVM) modeling using financial data. The main objective is to develop predictive models to forecast financial trends and prices over time.

Time series analysis is crucial for understanding and forecasting financial data. This project utilizes a dataset of historical financial data and applies SVM models to predict future trends. The project is structured to facilitate ease of understanding and replication of the results.

• Data Retrieval: Custom functions are used to fetch historical financial data from MetaTrader 5.
• Data Processing: The data is processed and transformed to prepare it for modeling, including steps such as standardization and feature engineering.
• Model Development: SVM models are developed and trained on the processed data.
• Model Evaluation: The performance of the models is evaluated using appropriate metrics to ensure robustness and accuracy.
• Visualization: Key results and trends are visualized to provide insights into the data and model performance.

• data.py: Contains functions for data retrieval and manipulation, including:

  • get_rates: Fetches historical price data for a given symbol and timeframe.
  • add_shifted_columns: Adds lagged versions of columns to a DataFrame to create features for time series prediction.
  • split_data: Splits data into training and testing sets.

• backtest.py: Provides functions for backtesting financial strategies, including:

  • compute_strategy_returns: Calculates returns for a given strategy.
  • plot_returns: Visualizes the cumulative returns of a strategy.
  • vectorize_backtest_returns: Computes and prints financial metrics such as Sortino, Beta, and Alpha ratios.
  • compute_model_accuracy: Calculates and visualizes the accuracy of model predictions.
  • strategy_drawdown: Computes and visualizes the drawdown of the strategy.

• SVR_Model.ipynb: Jupyter Notebook containing the main code for data processing, model development, evaluation, and visualization.

• .gitignore: Specifies files and directories to be ignored by Git.

• LICENSE: The MIT License under which this project is distributed.

• README.md: This file.

To run this project, you need to have Python installed on your machine along with the following packages:

• pandas
• numpy
• scikit-learn
• matplotlib
• MetaTrader5

1. Clone the Repository: Clone this repository to your local machine using git clone https://github.com/YarosInter/Time-Series-Support-Vector-Machine.git.

2. Install Dependencies: Ensure all required Python packages are installed.

3. Run the Notebook: Open SVR_Model.ipynb in Jupyter Notebook to explore the model and results.

4. Data Retrieval: Use data.py to fetch and preprocess financial data.

5. Backtesting: Leverage backtest.py to evaluate your strategy's performance.

Contributions are welcome! Please fork the repository and submit a pull request for any improvements or new features.

This project is licensed under the MIT License - see the LICENSE file for details.