The volume weighted average price execution model had its own implementation
of VWAP since the existing VWAP didn't perform the daily reset. The intraday
VWAP indicator has since been added to the indicators project so this is just
duplicate code.
Framework models should not loop over `algorithm.Securities` since it contains all securities that were ever added to the algorithm, but `UniverseManager.ActiveSecurities` that contains only the active securities.
Instances of `PortfolioTargetCollection` are intended to be a class level variables and not a method level variables. As a class member it maintains a complete set of all portfolio targets so you can operate against a 'full view' instead of the potentially streaming targets (which can come in one by one as alpha is generated).
- Use `MaximumDrawdownPercentPerSecurity` as `RiskManagementModel`.
- Modifies regression test to reflect risk model choice
- Use SetXXX to set models in python version
We are not using python lists instead of generator (yield) because we get better exception information in this case. The aim is to lead users to avoid using generators and/or know its limitations.
The exception in ignored because the generator isn't closed until it is being deleted (automatically in this case, when Python exits); the generator __del__ handler closes the generator, which triggers an exception of there is one.
We add entries for custom data w/ symbol references, so the generic look
up being performed here never resolves and throws an exception. With
custom data we add the entry manually and that same pattern has been
repeated here for consistency.
This happens when users pass Securities.Keys into the manual model,
causing the SecurityChanges object to have references to the canonical
securities, thereby leading to indicators and other things being
done to them unknowningly.
The PairsTradingAlphaModel is a simple example of defining an insight
grouping. Insights that are grouped together are assigned a unique
group-id that can be used by the portfolio construction model.
Updates were made to the CommonAlphaModelTests to give more control to
derived types. Some changes are still needed here to give securities
unique prices. I would recommend using a psuedo-random walk approach
by using Random with a constant seed value.
- Typoes were fixed in `ImmediateExecutionModel.cs`;
- Refactors `PriceIsFavorable` methods in `StandardDeviationExecutionModel` and `VolumeWeightedAveragePriceExecutionModel` C# models;
- Adds python version of C# execution models
The cleanest way to get a python type name, when we know it is an instance of a class, is getting the value of the `__class__.__name__` attribute.
- Changes `CommonAlphaModelTests.ModelNameTest`:
Since it requires some string manipulation to have the exact model name for C# and Python, we don't define `Name` in the python models and accept the default value from `AlphaModelPythonWrapper.Name`.
This was always logging, even when the risk model wasn't overriding
any of the targets. The new check checks to see if the risk model
created any override targets, and if so, we'll print all of the
targets as they will be submitted to the execution model.
The composite model combines multiple alpha models into a singular model and
properly sets each insight's SourceModel property to the name of the model that
generated the insight
Alpha models can choose to implement the Name property, if not, the system
will use the model's type name as the Insight.SourceModel.
Existing tests were updated to also assert expected model names
This identifier is used to determine the alpha model that generated.
This is NOT ideal, since it requires users to specify the value, more
thought will be givent to how we can resolve this value automatically
Adds test for surviving roundtrip copy operation.
The PyObject parameter is converted into selector function of `Func<T1,T2>` type so that the model can user universe selector methods defined in python algorithms.
AlexCatarino