Terra Algo Backtest

A package to replay trades and simulate strategies

• Documentation: https://terra_algo_backtest.readthedocs.io
• GitHub: https://github.com/terra-revival/terra_algo_backtest
• Free and open source software: Apache Software License 2.0

Features

• Load trades from CEX
• Load trades from DEX
• Support custom algo
• Replay trades against CEX
• Replay trades against DEX
• Display simulation metrics

Quick Start

1 - Install from source

$ git clone git://github.com/terra-revival/terra_algo_backtest
$ cd ./terra_algo_backtest
$ pip install -r requirements_dev.txt
$ make dist install
$ cd ./docs/examples
$ jupyter notebook

2 - Load trades from Binance

from binance import Client
from terra_algo_backtest.binance_loader import new_binance_client

# replace these with your Binance API key and secret
client = new_binance_client(
    os.getenv("BINANCE_API_KEY"),
    os.getenv("BINANCE_API_SECRET"))
# parameters
params = {
    "base_pair": "LUNC/BUSD",
    "quote_pair": "USTC/BUSD",
    "start": "2023-03-01 00:00:00",
    "end": "2023-06-28 23:59:59",
    "frequency" = Client.KLINE_INTERVAL_1HOUR,
}
# create trades
df_trades = client.create_trade_data(
    params["base_pair"],
    params["quote_pair"],
    params["vol_multiplier"],
    params["frequency"],
    params["start"],
    params["end"])

3 - Initialise a new market

from terra_algo_backtest.market import MarketQuote, new_market

# market liquidity
liquidity_usd = 100000
# LUNC/BUSD market price
base = MarketQuote(params["base_pair"], 0.00008)
# USTC/BUSD market price
quote = MarketQuote(params["quote_pair"], 0.01)
# LUNC/USTC market with 100,000$ liquity and 0.3% tx fee
mkt = new_market(liquidity_usd, quote, base, 0.003)

4 - Replay trades with your algo

from terra_algo_backtest.strategy import get_strategy
from terra_algo_backtest.simulation import swap_simulation

# simple DEX liquidity provider strategy
strategy = get_strategy("uni_v2")
# replay trades against DEX
simul = swap_simulation(mkt, df_trades, strategy)

5 - Visualise simulation results

from terra_algo_backtest.plotting import new_simulation_figure

# display results
show(new_simulation_figure(mkt, simul, plot_height=300))

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.