Skip to content

backtesting

John Kearney edited this page Sep 17, 2022 · 1 revision

Backtesting

If you want to see how a configuration will perform, you can run it through a backtest that is provided by Passivbot. While backtesting will not give you a guarantee on the performance for the future, it does give you valuable insight into how the bot would have performed in the past.

When you run the backtester, you should receive some output as it begins downloading the necessary price data from the exchange (using the API keys you provided earlier). The price data is cached on the machine and can be re-used between backtests and optimize sessions. This also means if you interrupt or close the process, it will continue downloading price data where it left off. The bot comes packaged with a downloader that allows the rapid retrieval of price data based upon provided dates, and works independently of the backtesting unit.

!!! Warning The bot runs backtests on trade data, making it as accurate as possible in backtesting. Be aware that other factors like latency and exchange connection issues can also play a role when you run a bot live.

Running a backtest

To execute a backtest, you can execute the following command from the root folder:

python3 backtest.py path/to/config_to_test.json

Running this command without any arguments other than path to live_config will make the backtester use the details provided in the configs/backtest/default.hjson file. When you first checkout the project, you will need to setup your exchange credentials in the api-keys.json (please read Running live for more details). The backtest needs a connection to the exchange to be able to download the trade data required for the backtest.

In the configs/backtest/default.hjson, the account name (specified in api-keys.json) needs to be provided. After updating the configs/backtest/default.hjson with your account name, you should be able to succesfully run a backtest.

Apart from the account name, there are a number of other parameters you can specify in the backtest configuration file:

  • the exchange
  • the account name (must match one specified in api-keys.json)
  • the symbol(s) to backtest on
  • the latency to simulate during backtesting
  • the starting balance
  • the start and end date for the backtest

Command-line arguments

Other than modifying the default.hjson file, it is also possible to specify a number of configuration options to use via the commandline. One or more arguments are provided to the backtester using the following syntax on the command line:

python3 backtest.py <key> <value>

The following options can be provided to the backtester. Note that any argument provided will override a value specified in the backtest configuration file.

Key Description
--nojit Disables the use of numba's just in time compiler during backtests
-b / --backtest_config The backtest config hjson file to use
Default value: configs/backtest/default.hjson
-d / --download-only Instructs the backtest to only download the data, but not dump the ticks caches to disk
-s / --symbol The symbol(s) to run the backtest on, separated by a ,
-u / --user The name of the account used to download trade data
--start_date The starting date of the backtest
Syntax: YYYY-MM-DD
--end_date The end date of the backtest
Syntax: YYYY-MM-DD
-bd / --base_dir the base directory to place the output files in
Default: backtests
-oh / --ohlcv use 1m ohlcv instead of 1s tick samples

Backtest results

When the backtest is completed, the results will be shown on the console. This includes things like average daily gain, maximum time that Passivbot was stuck etc. Apart from showing these results in the terminal, these results are also stored in backtests/{exchange}/{symbol}/plots/{datetime}/backtest_result.txt. This folder will also include the actual live_config.json file that was used for the plot, and several graphical plots. One of these for example is the balance_and_equity_sampled_{long/short}.png, which shows how the balance and equity evolved during the course of the backtest.

The file balance_and_equity.png will show how the balance and equity progressed during the period being backtested. The blue line in the graph represents the balance, and the orange line represents the equity.

Besides this file, whole_backtest_{long/short}.png and a number of backtest_XofY.png are also created.
The latter represent the entire backtest period, but are split up into separate files for easier inspection and zooming.

On these plots the blue dashed line is the long position price and the red dashed line is the short position price.
Blue dots are buys (long entries or short closes) and red dots are sells (short entries or long closes).
Red Xs are auto unstuck sells (long closes or short entries), blue Xs are auto unstuck buys (long entries or short closes)
and in static grid mode the secondary entry is a larger green dot.

The auto_unstuck_bands_{long/short}.png plots show the price thresholds at which auto unstucking orders would fill.
initial_entry_band_{long/short}.png shows the EMA limited initial entries.

Clone this wiki locally