GitHub

Running in Paper Mode

Paper mode provides a realistic simulation environment that uses live market data while simulating order execution. This allows you to test strategies with real market conditions without risking actual capital.

Configuration Options

In order to configure a strategy in paper mode, you can define the default mode in user/[planar.toml](../config.md#configuration)-file) or in your strategy project's Project.toml file. Alternatively, pass the mode as a keyword argument:

Configuration via TOML Files

# user/[planar.toml](../config.md#configuration)-file)
[Example]
mode = "Paper"
[exchanges](../exchanges.md) = "binance"
throttle = 5  # seconds between [strategy](../guides/strategy-development.md) calls
initial_cash = 10000.0
# Strategy Project.toml
[strategy]
mode = "Paper"
sandbox = true  # Use [exchanges](../exchanges.md) sandbox/testnet

Configuration via Julia Code

Starting Paper Mode

To start the strategy, use the following command:

Advanced Startup Options

Upon executing this, the following log output is expected:

Background Execution

To run the strategy as a background task:

The logs will be written either to the s[:logfile] key of the strategy object, if present, or to the output of the runlog(s) command.

Log Management

Comprehensive Setup Examples

Basic Spot Trading Setup

Advanced Margin Trading Setup

Multi-Exchange Paper Trading

Understanding Paper Mode

When you initiate paper mode, asset prices are monitored in real-time from the exchange. Order execution in Paper Mode is similar to SimMode, albeit the actual price, the trade amount, and the order execution sequence are guided by real-time exchange data.

Order Execution Mechanics

Market Orders

  • Market Orders are executed by surveying the order book and sweeping available bids/asks. Consequently, the final price and amount reflect the average of all the entries available on the order book.
  • Execution includes realistic slippage based on order book depth
  • Large orders may experience partial fills across multiple price levels

Limit Orders

  • Limit Orders sweep the order book as well, though only for bids/asks that are below the limit price set for the order. If a Good-Till-Canceled (GTC) order is not entirely filled, a task is generated that continuously monitors the exchange's trade history. Trades that align with the order's limit price are used to fulfill the remainder of the limit order amount.
  • Orders are queued and filled based on real market movements
  • Partial fills are handled realistically based on market liquidity

Real-Time Data Integration

Price Feeds

Order Book Integration

Performance Monitoring

Real-Time Metrics

See Also

Risk Management in Paper Mode

Position Sizing

Stop Loss Management