We are working on a new version of Gekko. I won't update the stable branch (the one you are looking at) anymore, all new functionality goes into the new version. The new version has a lot more capabilities:
- A datastore that persists data to disk so we don't rely on either exchange APIs providing historical data (almost all exchanges are dropping this support) or third parties (like bitcoincharts). This means support for altcoin markets :).
- A new actor system that makes it trivial to add new modules to Gekko that respond to events (market events or trading advice).
- A rewrite of basically all internals.
- This time with a test suite :)
- An API for trading methods.
All the code is located in the localDB branch, most of the discussion happened here and on IRC (#gekkobot on freenode). Here is the todo list before it's done.
After that we are thinking about:
- webbased GUI.
- More trading methods (using talib for example).
- Use a datastore of your choice (like a database etc).
Join the discussion and help out!
The point is ladies and gentlemen that greed, for lack of a better word, is good.
-Gordon Gekko
Gekko is a Bitcoin trading bot for popular exchanges written in javascript running on nodejs, it will feature multiple trading methods using technical analysis (at this moment it only knows EMA). It can monitor the live market or backtest strategies on historical data.
- Paper trading
- Live trading (trade bot)
- Backtesting
This project is a learning excercise of me, a student with some experience in programming (mostly web) and zero experience in economics and trading. I figured writing my own trade bot would be the best way to learn about implementing mathematical trading algorithms. So here is my very first attempt at anything related to trading / algorithmic decision making.
I'm developing Gekko fully open source in the hope of getting feedback from folks with more experience in this field. Because I not only want to attract programmers I am doing my best to make the code as readable as possible, this includes a lot of comments and probably not the most efficient (but expressive) code.
As this is a learning experience for me all feedback is extremely appreciated. If you don't want to contribute to the code you can always just send me an email or leave feedback in the Gekko thread on the bitcointalk forum.
Use Gekko at you own risk.
Gekko uses Technical Analysis (only EMA at this moment) to determine when to enter or leave the live market, at such a moment it can:
- Advice on what you should do (for example per email).
- Calculate the profits of all the advices so far.
- Automatically trade.
Gekko currently supports paper & live trading at the following exchanges:
Gekko supports paper trading at the following exchanges:
- Bitstamp (Live trading broken at the moment)
You can also backtest strategies using Gekko. I am currently expanding and improving the backtester and I could use all the feedback I can get. Keep in mind that I currently would not suggest applying strategies for real investments based on the result, consider it alpha stage.
You can find more detailed information in the document about Backtesting as well as how to set it up.
Windows user? Here is a step-by-step guide on how to get Gekko running on Windows.
Gekko runs on nodejs, once you have that installed you can either download all files in a zip or clone the repository via git:
git clone git://github.com/askmike/gekko.git
cd gekko
You need to download Gekko's dependencies, which can easily be done with npm:
npm install
To change the settings, open up and edit config.js to change the parameters. Check out doc on Configuring Gekko where everything is explained in more detail.
To run the bot you just have to start Gekko:
node gekko
You can also run Gekko silently, for examples on how to do this check out the advanced features.
If you installed the bot via git you can easily fetch the latest updates by running:
git pull && npm update
If you started Gekko it will remain open in your terminal and log out new information, for example:
start time: 2013-05-19 23:17:38
I'm gonna make you rich, Bud Fox.
Let me show you some Exponential Moving Averages.
2013-06-02 18:21:15 (INFO): ADVICE is to HOLD @ 117.465 (0.132)
2013-06-02 18:21:15 (INFO): (PROFIT REPORT) original balance: 207.465 USD
2013-06-02 18:21:15 (INFO): (PROFIT REPORT) current balance: 217.465 USD
2013-06-02 18:21:15 (INFO): (PROFIT REPORT) profit: 10.000 USD (4.820%)
After the first fetching, every new interval (in the config) Gekko will fetch new trade data, advice on what to do and give a profit report:
- HOLD means don't do anything, we are either not in a trend or the trend has not changed since last check.
- BUY means the trend has changed to an uptrend, advice is to buy now so we can sell at the end of the trend.
- SELL means the trend has chacnged to a downtrend, advice is to sell now so we can buy back at the end of the trend.
After every line of advice we can see the current price Gekko calculated and the difference in EMAs, this makes it easier to understand the advice.
The profit report will log out Gekko's profit since it started, this is done using a buy and sell simulations (regardless if you have automatic trading enabled or not). Gekko applies the configured trading fee on both simulated sells and buys.
If Gekko logs 20% that means that if you would have had automatic trading enabled on an exchange account with a balance of 1BTC, you would now have 1.2BTC.
If you configured Gekko to automatically sell on this information it will also log:
- NOW going to BUY, when it is buying BTC.
- NOW going to SELL, when it is selling BTC.
- Add more exchanges (such as bitfloor, campBX).
- Store trade history in levelDB.
- Add money management (so that the bot will only spend a start amount of X and the profits on top of X).
- The title is inspired by Bateman.
- This project is inspired by the GoxTradingBot Chrome plugin (though no code is taken from it).
If Gekko helped you in any way, you can always leave me a tip at (BTC) 13r1jyivitShUiv9FJvjLH7Nh1ZZptumwW
The MIT License (MIT)
Copyright (c) 2013 Mike van Rossum [email protected]
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent
