Risk Management on specific condition met

Hi Axist,

When using a risk management model, the risk logic usually goes in the ManageRisk method of the framework module. LEAN automatically calls this method, so we don't need to manually call it from the algorithm class.

However, to call a method of the risk management model, save a reference to the model in the initialize method.

# In Initialize
self.risk_model = BracketRiskModel(self.down_value, self.up_value)
self.SetRiskManagement(self.risk_model)

# In OnData
self.risk_model.my_method()

Best,
Derek Melchin

Want to invest in QuantConnect as we build the Linux of quant finance? Checkout our Wefunder campaign to join the revolution.

Thanks for this, that makes sense.  What if I wanted to exempt securities or invested holdings from the self.SetRiskManagement(self.risk_model) ?

Say for example something like this in the following trailing stop risk management model which is being called from in the main.py file.  This following code will error because the ‘Securities’ object is not recongized despite the fact that there's a reference to the “from main import *” file at the top.

#region imports
from AlgorithmImports import *
from main import *
#endregion
from QuantConnect.Algorithm.Framework.Risk import RiskManagementModel


# bracket risk model class
class BracketRiskModel(RiskManagementModel):
    '''Creates a trailing stop loss for the maximumDrawdownPercent value and a profit taker for the maximumUnrealizedProfitPercent value'''
    def __init__(self, midDrawdownPercent = 0.025 ,maximumDrawdownPercent = 0.1, midUnrealizedProfitPercent = 0.025, maximumUnrealizedProfitPercent = 0.1):
        #it was maximumDrawdownPercent maximumUnrealizedProfitPercent  .05
        self.maximumDrawdownPercent = -abs(maximumDrawdownPercent)
        self.midDrawdownPercent = -abs(midDrawdownPercent)
        self.trailingHighs = dict()
        self.maximumUnrealizedProfitPercent = abs(maximumUnrealizedProfitPercent)
        self.midUnrealizedProfitPercent = abs(midUnrealizedProfitPercent)


    def ManageRisk(self, algorithm, targets):
        riskAdjustedTargets = list()
        for kvp in algorithm.Securities:
            symbol = kvp.Key
            security = kvp.Value


            # Remove if not invested -
            ''' BUT EXEMPT THE SPY FROM BEING ASSOCIATED WITH THE FOLLOWING TRAILING STOP CALCULATIONS '''
			#
            if not security.Invested and self.Securities['SPY'].Invested:
                self.trailingHighs.pop(symbol, None)
                continue
            pnl = security.Holdings.UnrealizedProfitPercent
           
            if pnl > self.maximumUnrealizedProfitPercent:
                # liquidate
                algorithm.Debug(f"Profit Taken: {security.Symbol}")
                algorithm.Log(f"Profit Taken: {security.Symbol}")
                riskAdjustedTargets.append(PortfolioTarget(security.Symbol, 0))
                return riskAdjustedTargets
               
            # Add newly invested securities
            if symbol not in self.trailingHighs:
                self.trailingHighs[symbol] = security.Holdings.AveragePrice   # Set to average holding cost
                continue


            # Check for new highs and update - set to tradebar high
            if self.trailingHighs[symbol] < security.High:
                self.trailingHighs[symbol] = security.High
                continue


            # Check for securities past the drawdown limit
            securityHigh = self.trailingHighs[symbol]
            drawdown = (security.Low / securityHigh) - 1


               
            if drawdown < self.maximumDrawdownPercent:
                # liquidate
                algorithm.Debug(f"Losses Taken: {security.Symbol}")
                algorithm.Log(f"Losses Taken: {security.Symbol}")
                riskAdjustedTargets.append(PortfolioTarget(symbol, 0))
               
        return riskAdjustedTargets

Actually found the answer I was looking for above here, nevermind!

Had to do with how to refer to self.securities like value when dealing with:

for kvp in algorithm.Securities:
    symbol = kvp.Key
    security = kvp.Value