Skip to content

Latest commit

 

History

History
101 lines (64 loc) · 3.48 KB

README.md

File metadata and controls

101 lines (64 loc) · 3.48 KB

cryptrade

Cryptrade is a cryptocurrency market visualization and automated trading program written in python, with the aim of supporting collection and plotting of live trade data, batch backtesting of trading strategies, live trading simulation, and automated live trading. Current development is focused on the Bitfinex exchange.

This (pre)release includes a very limited implementation:

Live monitoring and visualization of the Bitfinex BTC/USD market
Automated live trading (experimental) of the Bitfinex BTC/USD market
Moving average crossover trading strategies with stoploss and risk management

For now only the Bitfinex API is supported.

A batch backtesting framework is in progress.

Dependencies

NumPy, MatPlotLib, Requests, and PyQt4

Getting Started

An installer is not yet implemented; scripts can be run from the same directory as the cryptrade package, or the cryptrade package can be placed in the python path. A data directory will be created where the scripts are run if they are set to record trade and candle data.

Example 1 demonstrates how to start a simple live trade monitor that records and prints new trades in the terminal window as they come in from the exchange.

Example 2 demonstrates how to start the GUI, which provides a candle plot and record of the most recent trades. The GUI currently supports visualization of moving average and MACD indicators, as well as a very experimental implementation of a live trade bot based on a moving average crossover strategy. I AM IN THE EARLY STAGES OF TESTING THE TRADE BOT; I DO NOT RECOMMEND USING IT AT THIS TIME UNLESS WITH NEGLIGIBLE AMOUNTS OF MONEY.

On the first run with no local data available, the most recent 2 hours of trades will be obtained from the API. If set to record data (the default case), local files of trade and candle data will be accumulated as the program continues to run.

GUI Usage

Market view window

Upon starting the GUI, the market view window will display a candle graph of recent prices. Individual trades are listed at the bottom.

Adding indicators

The "Add indicators" button opens the indicator dialog where new indicators can be selected. Simple and exponential moving averages (SMA and EMA) and moving average convergence-divergence (MACD) indicators are currently implemented.

Market view with indicators

Plots of indicators can be toggled on and off from the market view window from the list in the lower-right corner.

Setting up trading

The "Trader setup" button opens a dialog where a trading strategy is selected (currently only a Moving Average Crossover strategy is implemented). If an API key was not provided when the gui was initialized, it should be entered here to begin live trading (still in testing and not recommended!).

Trader window

The trader window draws from the API to list personal trade history, current open orders, and current finances.

Acknowledgements

Plot design is based on a tutorial series from Sentdex.com

http://sentdex.com/sentiment-analysisbig-data-and-python-tutorials-algorithmic-trading/how-to-chart-stocks-and-forex-doing-your-own-financial-charting/

Bitfinex API python implementation is based on sample code by Raphael Nicolle

https://community.bitfinex.com/showwiki.php?title=Sample+API+Code