- Fixes for C# RegisterIndicator API methods which were ignoring provided
type of T
- Fixes for Py RegisterIndicator API methods which was not using the
provided 'selector' method
- Adding C# and Py regression algorithm
Update test to the new crypto and equity subscriptions rules:
- Only consolidates trades
- Low resolution data are only trades.
Update and fix tests
Add missing minute sample files
Update Regression algorithms statistics
Crypto and Equities will consolidate trades by default
Daily and hourly resolution will return only trades for crypto and equities.
Update equity TAQ regression test
Add minute sample files
Updating regression algorithms
Update SpotMarket test cases
Add regression test for equity trades and quotes
- History request.
- Trades and quotes pumped into OnData.
- Subscriptions are added correctly.
Add sample data
Checks low resolution only subscribes to trade bars
Implements `PortfolioBias` in EWPCM, CWPCM and IWPCM. With this new feature, these PCM will ignore insights that do not respect the desired bias. E.g. for `PortfolioBias.Long`, on Insights with `InsightDirection.Up` will be converted into `PortfolioTarget.Quantity` greater than zero and other `InsightDirection` will result in `PortfolioTarget.Quantity` of zero.
- Adding new unit test for python PCM implementations, asserting each
method is correctly called
- Reverting some unrequired changes in the
`MeanVarianceOptimizationFrameworkAlgorithm`
- Refactor shared logic from `EqualWeightingPortfolioConstructionModel`
into base `PortfolioConstructionModel` implementation
- `MeanVarianceOptimizationPortfolioConstructionModel` will respect
rebalancing period and will use all active inisights, not just the last
- Adding new `Func<DateTime, DateTime?>` that allows PCM to return null
if the next rebalance time is null, in which case the function will be
called again in the next loop.
- Adjusting PCM next rebalance time check to perform rebalance once the
time is reached
- Adding new regression test. Updating existing
- Moving InsightCollection into base `PortfolioConstructionModel`
- Will call `InsightCollection.GetNextExpiryTime()` on each check, and
for performance `InsightCollection` will keep track of next insight
expiry time
- Removing need for PCM base classes having to call `RefreshRebalance`
- Some refactor clean up at base
PortfolioConstructionModel.IsRebalanceDue()
- Refactoring some PCM methods to be `protected` since they are not required
to be public
- Adding new `PortfolioConstructionModel.RebalanceOnInsightChanges`
flag, that will allow avoiding new insights or insight expirations to
trigger a rebalance
- Updating unit tests
- Fix for the MeanVarianceOptimizationPortfolioConstructionModel that
was skipping, in some cases, 0 magnitude insights
- Improving OrderSizing.Value and Volume to include code in consumers
- OrderSizing.GetUnorderedQuantity() will adjust result by lot size
- ImmediateExecutionModels will use OrderSizing.GetUnorderedQuantity()
- OrderSizing.Value() will take into account ContractMultiplier
- Adding unit tests
- Add missing PCM constructor methods for the different supported
rebalancing periods overloads
- Normalize rebalance behavior in the base `PortfolioConstructionModel`
- Adding new `PortfolioConstructionModel.RebalanceOnSecurityChanges`
that will allow disabling rebalance on security changes
- Adding unit tests
Changes `GetLastKnownPrice` logic to retry to get non-null data after a first attempt. Previously, it would return null in the first attempt and illiquid securities would not have valid data to set its market price. In the second attempt, we increase the look-back period to the equivalent of three trading days worth of data.
- `AddUniverse` call will add new Universe to the pending collection which
will be consumed at the `OnEndOfTimeStep` where it will be added to the
data feed, same as we do for the `UserDefinedUniverses`. This is
required since the start and end date, during initialize, is consumed by
these universe subscriptions.
-Adding unit test.
The class field that tracks the current month is updated only if there are securities that passed the selection criteria. It prevents division by zero and allows the universe selection a new attempt on the next trading day while keeps the universe unchanged
Since extension methods don't play well with pythonnet, this change converts
the extensions class into a decorator class. Additionally, this ObjectStore
type is the type that gets exposed via QCAlgorithm so users can access these
methods directly without requiring the use of extension methods.
This approach has many good properties. For one, it doesn't force implementors
of IObjectStore to use a base class. Second, it maintains healthy separation of
API level concerns (such as convenient methods) from the abstraction level conerns
of IObjectStore. Setting it up in this way ensures ANY implementation of IObjectStore
will still get access to these additional methods. Another thing to note is this
prevents using a base class on QCAlgorithm's public interface. Instead, we have a
specific type that is dedicated to fulfilling API level requirements, which also
provides us flexibility in the event the API needs to be updated. If it were a subclass,
you run the risk of breaking the implementors of the subclass.
This commit is squashed from iterative development:
- More consistent method naming
- Storage root path updated to be absolute and include algorithm name
- Storage root path created only if object store is actually used
- Implemented XML save/load
- Added missing unit tests
- Replaced Log.Trace with Log.Error calls
- Added the object store name logging in Engine.Main
- Read storage root from config
- Create algorithm storage root folder in Initialize
- Remove empty folder in Dispose
- Added null checks in all methods
- Added missing XML parameter docs
- make Initialize and Dispose virtual
- make AlgorithmStorageRoot protected
The IObjectStore abstraction provides algorithms with a persistent
storage mechanism. While the algorithm is running, data is maintained
in memory as a dictionary of raw bytes (string -> byte[]). This ensures
we avoid any reference type shenanigans. Periodically, the data in the
object store is persisted and additionally, when the algorithm shuts
down, the object store's data will again be persisted. This ensures that
when the algorithm starts up again, it will have access to any state
that has been saved into the object store.
A great use case for IObjectStore is saving a compute heavy model.
For example, computing the weights of a deep neural network is very
CPU intensive, but after the weights are computed, evaluation is fairly
quick. An initial backtest can be used to solved for the network's weights
and then subsequent backtests or even in live mode, the weights will be
available to the algorithm provided they were saved into the object store.
Also, some libraries require a file path to load model data. The object
store provides a `GetFilePath(key)` method which will copy the data for
the provided key to the disk and return that path so the library can load
the model data.
- Renaming `PreSelected` to `Constituents`
- Adding base `ConstituentsUniverse`
- Adding Py and C# regression algorithm
- Fixing bug in `UniverseSelection`, it wasn't removing pending to be
removed securities unless the universe selection changed
- Adding test data
- `SetHoldings` will take `OnMarketOpen` ordes into account when
determining order quantity
- Adding new regression test. Updating existing algorithms which
suffered of the issue
- Adding a performance improvement, will avoid margin and portfolio
calculations for MarketOnOpen orders that wont be able to fill
Martin-Molinero