Using the BacktestController Class

The BacktestController class is responsible for running backtests on your chosen investment strategy and saving the results.

A step-by-step guide follows for instantiating this class:

Steps to Instantiate

Prepare your Investment Strategy:

Before instantiating BacktestController, you should already have an instance of your desired investment strategy:

import investos as inv

strategy = inv.portfolio.strategy.SPO(
    actual_returns = actual_returns,
    forecast_returns = forecast_returns,
    costs = [
        ShortHoldingCost(
            short_rates=short_rates,
            exclude_assets=["cash"]
        )
    ],
    constraints = [
        MaxShortLeverageConstraint(limit=0.3),
        MaxLongLeverageConstraint(limit=1.3),
        MinWeightConstraint(),
        MaxWeightConstraint(),
        LongCashConstraint()
    ],
    cash_column_name="cash"
)

Instantiate the BacktestController Class:

Use the strategy instance from the previous step as the first argument.

backtest_controller = inv.portfolio.BacktestController(
    strategy=strategy
)

Provide Optional Arguments (if necessary):

• hooks: A dictionary of hooks for specific events during the backtest. Currently, the only hook that exists is after_trades. You can use it to do things like increase / decrease leverage based on performance, change constraints on the fly, print output on each iteration, etc.

hooks = {
    "after_trades": [
        lambda backtest, t, u, h_next: print(".", end=''),
    ]
}

• initial_portfolio: A pandas.Series indicating the initial portfolio allocation. If not provided, the controller will create a default initial portfolio with an AUM of 100,000,000 (allocated to cash) and no other initial allocations.

• results_model: A custom result class to store backtest results. It defaults to BaseResult if not provided.

• time_periods: A series of time periods you wish to backtest. By default, it uses actual_returns.index from your investment strategy instance.

• start_date and end_date: Start and end dates for the backtest. It uses the first and last dates from time_periods if not set.

• aum: The initial portfolio's asset under management. Defaults to $100MM.

Example instantiation with optional arguments:

backtest = BacktestController(
    strategy=strategy,
    start_date='2020-01-01',
    end_date='2021-01-01',
    aum=50_000_000,
    hooks={
        'after_trades': [custom_hook_function]
    },
)

Run the Backtest:

Once you've instantiated BacktestController, you can run the backtest using generate_positions.

This returns an instance of BaseResult for performance reporting.

results = backtest.generate_positions()
# result.summary will print high-level results

Conclusion

With the steps mentioned above, you can effectively instantiate the BacktestController class and execute a backtest with your desired strategy.

Adjust the optional parameters according to your needs to fine-tune your backtesting process!

Next: Investment Strategies

Next, let's explore the off-the-shelf investment strategies available to you: Off-The-Shelf Investment Strategies.