Intraday breakout alpha model

To help you develop a robust alpha model for intraday stock breakout detection using a volume spike condition and moving average crossover, here's a structured approach based on QuantConnect's framework:

Step-by-Step Guide

1. Define the Universe Selection:

   • Use CoarseFundamentalUniverseSelectionModel to filter stocks based on criteria such as liquidity or market cap.
   • Example:

     def coarse_selection_function(self, coarse):
         sorted_by_dollar_volume = sorted(coarse, key=lambda x: x.dollar_volume, reverse=True)
         return [x.symbol for x in sorted_by_dollar_volume[:50]]
     

2. Initialize Indicators:

   • Set up short-term and long-term EMAs and a volume spike indicator for each stock.
   • Example:

     self.short_ema = self.ema(symbol, 10, Resolution.MINUTE)
     self.long_ema = self.ema(symbol, 30, Resolution.MINUTE)
     

3. Implement the Alpha Model:

   • Create a custom Alpha Model that generates insights when a moving average crossover occurs alongside a significant volume spike.
   • Example:

     class BreakoutAlphaModel(AlphaModel):
         def update(self, algorithm, data):
             insights = []
             for symbol, symbol_data in self.symbol_data_by_symbol.items():
                 if symbol_data.is_breakout():
                     insights.append(Insight.price(symbol, timedelta(minutes=30), InsightDirection.UP))
             return insights
     

4. Handle Security Changes:

   • Use the on_securities_changed event to initialize indicators for newly added securities.
   • Example:

     def on_securities_changed(self, algorithm, changes):
         for security in changes.added_securities:
             self.symbol_data_by_symbol[security.symbol] = SymbolData(algorithm, security.symbol)
     

5. Update Method:

   • In the update method of the Alpha Model, check for the crossover and volume spike conditions.
   • Example:

     def is_breakout(self):
         return self.short_ema.current.value > self.long_ema.current.value and self.volume_spike()
     

6. Backtest and Optimize:

   • Run backtests to evaluate the performance and optimize parameters like moving average periods and volume spike thresholds.

7. Deploy and Monitor:

   • Deploy the strategy in a live environment and monitor its performance, making adjustments as necessary.

Key QuantConnect Methods

• AddEquity: To add equities to your algorithm.
• SetAlpha: To set your custom Alpha Model.
• OnSecuritiesChanged: To handle changes in the universe.
• SetUniverseSelection: To define how securities are selected.
• SetStartDate: To set the start date for your algorithm.

This approach should help you create a robust intraday breakout detection model. Adjust the parameters and logic to fit your specific needs and market conditions.

Sources: Key Concepts - QuantConnect.com, Head & Shoulders TA Pattern Detection, Legacy Fundamental Universes - QuantConnect.com, Price Gap Mean Reversion Alpha, V I X Dual Thrust Alpha