Universe Selection

In QuantConnect, the add_universe method is used to dynamically select stocks for your algorithm using a two-step process: coarse and fine selection. Here's how they work together:

1. Coarse Selection: This is the initial filtering step. You define a coarse selection function that takes a list of CoarseFundamental objects and returns a list of Symbol objects. The coarse selection is typically used to filter stocks based on basic criteria such as dollar volume and whether the stock has fundamental data. This step is lightweight and helps reduce the number of stocks for further analysis.

   Example of a coarse selection function:

   def coarse_selection_function(self, coarse: List[CoarseFundamental]) -> List[Symbol]:
       # Filter stocks with fundamental data and sort by dollar volume
       selected = [c for c in coarse if c.has_fundamental_data]
       sorted_by_dollar_volume = sorted(selected, key=lambda c: c.dollar_volume, reverse=True)
       return [c.symbol for c in sorted_by_dollar_volume[:100]]  # Top 100 by dollar volume
   

2. Fine Selection: After the coarse selection, the fine selection function further narrows down the list using detailed fundamental data. This function takes a list of FineFundamental objects and returns a list of Symbol objects. Fine selection allows you to filter stocks based on more specific criteria such as earnings, P/E ratio, or other fundamental metrics.

   Example of a fine selection function:

   def fine_selection_function(self, fine: List[FineFundamental]) -> List[Symbol]:
       # Sort by P/E ratio and select top entries
       sorted_by_pe_ratio = sorted(fine, key=lambda x: x.valuation_ratios.pe_ratio, reverse=True)
       return [x.symbol for x in sorted_by_pe_ratio[:10]]  # Top 10 by P/E ratio
   

3. Coordination: In the initialize method of your algorithm, you use add_universe to coordinate these two functions. The method first applies the coarse selection function to filter the initial list of stocks and then applies the fine selection function to the output of the coarse selection.

   Example of using add_universe:

   class MyUniverseAlgorithm(QCAlgorithm):
       def initialize(self) -> None:
           self.set_start_date(2020, 1, 1)
           self.set_end_date(2020, 2, 1)
           self.universe_settings.resolution = Resolution.DAILY
           self.add_universe(self.coarse_selection_function, self.fine_selection_function)
   

This setup allows your algorithm to dynamically select stocks based on your defined criteria, focusing on strategy development rather than static stock selection.

Sources: Initialization - QuantConnect.com, Legacy Fundamental Universes - QuantConnect.com, Key Concepts - QuantConnect.com, Legacy Fundamental Universes - QuantConnect.com, Stock Selection Strategy Based On Fundamental Factors

Ok. I understand that universe selection takes two steps: coarse and fine selections. The coarse selection returns list of symbol objects but I do not see where they are passed to the fine selection function as a starting point for further narrowing down the list.

Can you explain this “After the coarse selection, the fine selection function further narrows down the list using detailed fundamental data. This function takes a list of FineFundamental objects and returns a list of Symbol objects.”?