https://github.com/ata-turhan/vbt-backtester

https://github.com/ata-turhan/vbt-backtester

Last synced: 11 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/ata-turhan/vbt-backtester
• Owner: ata-turhan
• License: gpl-3.0
• Created: 2023-11-23T15:33:59.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-12-28T17:44:16.000Z (about 1 year ago)
• Last Synced: 2024-04-09T01:04:20.118Z (9 months ago)
• Language: Jupyter Notebook
• Size: 19.6 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.MD
  • License: LICENSE

Awesome Lists containing this project

README

        
# Overview

This Jupyter Notebook utilizes various Python packages, including numpy, pandas, numba, vectorbt, and optuna. The notebook is structured to perform the following tasks:

## 1. Getting Price Data

   - In this section, we fetch OHLCV data for different assets with varying time intervals.

## 2. Creating Signal

   - An example signal is generated using fast and slow moving averages, along with RSI values.

## 3. Simple Backtest

   - A straightforward backtest is conducted with a basic configuration. The results include detailed statistics, various plots, and a transaction table for analysis.

## 4. Window Optimization

   - Optimal values for slow and fast moving averages are determined using the built-in grid search functionality of the vectorbt package.

## 5. Trailing Stop Loss & Take Profit

   - Custom functions are incorporated to set stop loss and take profit values.

## 6. FPE Box

   - The logic for the false positive elimination box is implemented in this section.

## 7. Grid Search

   - An ordinary grid search is performed for hyperparameter optimization.

## 8. Bayesian Search

   - A Bayesian search is executed for hyperparameter optimization using the optuna package.




Feel free to run the notebook and explore the various analyses and optimizations provided. For additional details, refer to the code and comments within the notebook.