How to make Set SlippageModel and FeeModel work?

It seems like you are trying to set a custom slippage and fee model for options trading. However, the order fee and fill price are not reflecting your custom settings.

The issue might be related to the way you are setting the slippage and fee models. In QuantConnect, you should set the slippage and fee models on the security object, not on the option object.

Here is an example of how to set a custom slippage and fee model:

class MySlippageModel:

 def get_slippage_approximation(self, asset: Security, order: Order) -> float:
 return asset.price * 0.0001 * np.log10(2*float(order.absolute_quantity))

class MyFeeModel(FeeModel):

 def GetOrderFee(self, parameters):
 fee = max(1, parameters.Security.Price - parameters.Order.AbsoluteQuantity - 0.00001)
 return OrderFee(CashAmount(fee, 'USD'))

# In the Initialize method, set the slippage and fee model
security.set_slippage_model(MySlippageModel())
security.set_fee_model(MyFeeModel())

In your case, you should replace MySlippageModel and MyFeeModel with your custom models and set them on the security object.

Please note that the slippage model helps to calculate the order quantity and the fee model helps to calculate the order fee. However, the algorithm doesn't update the cash book with the fee from the fee model. The algorithm uses the actual fee from the brokerage to update the cash book.

For more information, please refer to the QuantConnect documentation and QuantConnect forum discussions.

Hi Mia, nice talking to you again ;)

But, in the case above, self.option is a security object. (To my understanding) 

Hi Kryz,

You're correct that self.option is a security object, specifically it's an Option object. However, the slippage and fee models should be set on the specific security objects that are part of your portfolio, not on the Option object itself.

In QuantConnect, self.Portfolio contains SecurityHolding objects. It is a dictionary of Symbol → SecurityHolding. So when you iterate over it, depending on how you do it, you either iterate over Symbol objects, SecurityHolding objects or tuples of (Symbol, SecurityHolding). The SecurityHolding can be of any class (ie equity, forex, crypto, options etc.).

Therefore, you should set the slippage and fee models on the specific securities that you are trading. For example, if you are trading AAPL options, you should set the models on the AAPL option security objects.

Here is an example of how you might do this:

# Add the option
option = self.AddOption("AAPL", Resolution.Minute)

# Set the filter
option.SetFilter(self.OptionFilter)

# Get the security object for the option
security = self.Securities[option.Symbol]

# Set the slippage and fee models on the security object
security.SetSlippageModel(ConstantSlippageModel(0.05))
security.SetFeeModel(ConstantFeeModel(1.62))

I hope this helps! Let me know if you have any other questions.

For more information, please refer to the QuantConnect documentation and QuantConnect forum discussions.

Can you define in Initialize method the slippage for all options?

Yes, you can define the slippage for all options in the Initialize method by using the SetSecurityInitializer method. This method allows you to set a custom security initializer that will be applied to all securities added to your algorithm.

Here is an example of how you might do this:

class MyAlgorithm(QCAlgorithm):

    def Initialize(self):
        self.SetStartDate(2024, 3, 1)
        self.SetEndDate(2024, 4, 30)
        self.SetCash(100000)

        self.option = self.AddOption("AAPL", Resolution.Minute)
        self.option.SetFilter(self.OptionFilter)

        self.SetSecurityInitializer(self.CustomSecurityInitializer)

    def CustomSecurityInitializer(self, security):
        security.SetSlippageModel(ConstantSlippageModel(0.05))
        security.SetFeeModel(ConstantFeeModel(1.62))

In this example, the CustomSecurityInitializer method sets the slippage and fee models for each security. This method will be called for each security added to your algorithm, including all options.

For more information, please refer to the QuantConnect documentation and QuantConnect forum discussions.

This works. Thx!

You're welcome, Kryz! I'm glad to hear that the solution worked for you. If you have any more questions or need further assistance, feel free to ask. Happy coding!