nash-makerbot 0.1.5

A crypto market maker bot that is easy understand and customize

The makerbot is a cryptocurrency market-maker bot that is easy to understand and customize. Its goal is to create a useful practical tool to begin trading digital assets algorithmically. The bot provides a skeleton to build upon for traders wanting to perform market-maker strategies. Its default configuration should allow for trading on Nash out of the box, so users can start quickly.

Requires Python 3.7

This software is a beta. It works as intended but several improvements are expected in the near term.

Quick install guide

First you will need a Nash (https://nash.io) account with 2FA disabled. The login credentials will be asked after starting the bot.

Install a Python 3 package manager if you do not already have one:

• On Ubuntu/Debian: sudo apt install python3-pip

• On Fedora/CentOS: sudo dnf install python3-pip

• On Mac OSX, first install Homebrew, then Python 3:

  • /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

  • brew install python3

• Using pip, install the bot:

  • as a user: pip3 install --user nash-makerbot

  • as root: pip3 install -U nash-makerbot

• Get the example configuration file provided here: $ wget https://gitlab.com/nash-io-public/nash-makerbot/raw/master/default.ini

• Update your $PATH to include the makerbot command by either re-entering the terminal or:source ~/.bashrc

• Start the bot: makerbot start eth_usdc --config=default.ini

How it works

This bot implements a simple symmetric market-maker algorithm. It is based on the idea that there is intrinsic volatility within a market, as shown in the image below, and will trade within a fixed spread.

When the bot starts, it watches the order book for the chosen market until it gets min_history_points updates or a maximum time of max_loading_time passes. The order book is stored in the OrderBookSeries object. This is a NamedTuple with two fields wrapping around two NumPy.Array s, one obs.t that contains the time in which each snapshot of the order book was recorded and another obs.data that contains the snapshots of the order book.

1. The bot will place the first buy order, the scrum buy. It is defined as setup_scrum_buy.

2. If the price goes up, the scrum buy order will be filled. In that case, the bot will place a new buy order buy 1 with a price buy_down_interval lower than the microprice.

3. At the same time as placing buy 1, a sell order corresponding to the scrum buy sell (sb) will be placed at a price straddle higher than buy 1. The bot has now become a market maker, since it has both buy and sell orders open in the limit order book.

4. As the market continues to move, the bot will place buy and sell orders according to its volatility. For each new buy order, a corresponding sell order will be placed.

Users are invited to take a closer look at the functions the bot follows when placing its orders: should_rebuy, should_place_buy, get_buy_order and size_order.

• should_rebuy determines if the bot should cancel the current open buy order. This is needed, for example, if the spread between the lowest sell and the current buy is too big or if the market has moved up and the probability of the buy order executing is now low.

• should_place_buy looks at market microstructure and determines if now is a good moment to place a new buy order.

• get_buy_order determines the price of buy orders.

• size_order determines the size of buy orders.

Todo

• [ ] Refactor to make code more pythonic and clean

• [ ] Add testing everywhere

• [ ] Generalize for custom strategy

• [ ] Create web UI

• [ ] Add historical data collection

• [ ] Allow simulation of a strategy and track expected performance.

License

This software uses the MIT License. You are free to use it for profit or include it in other closed-source projects.

History

0.1.5 (2019-12-12)

• Fix top bid check

0.1.4 (2019-12-08)

• Adjust timings

• Make retry use pg backoff

• Respect min base amount on orders

0.1.3 (2019-12-06)

• Fix typos

0.1.1 (2019-12-06)

• Add basic documentation

• Bump nash-api requirement to 1.0.6 to allow LINK trading

• Remove some dead code

0.1.0 (2019-12-04)

• Public release