* Fix bug, add unit tests and nit changes
The bug was raised because if two inputs points with repeated close values were inserted,
and the second one had 0 volume, when the first one is removed from
_oldDataPoints, as the reamaining value of it in _volumePerPrice is
zero, that entry is removed from _volumePerPrice. Therefore, when
the second input point is removed from _oldDataPoints, this entry
is not found in _volumePerPrice. Then the changes made were:
- Modify MarketProfile.Add() method to check every removed data point is
still in _volumePerPrice
- Add unit tests to covert this change
- Nit changes in VolumeProfileTests.cs
* Fix bugs and add more unit tests
- The Fill Forward bars shouldn't be considered in the calculations as
it's not real data.
- If bar inserted is not FF but has zero volume, it could remove a price
value from VolumePerPrice and leave this sorted list empty, then each
time we access that list we need to first check it's not empty
* Enhance ComputeNextValue() and add more unit tests
Regression Tests / build (push) Has been cancelled
Benchmarks / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
Research Regression Tests / build (push) Has been cancelled
Python Virtual Environments / build (push) Has been cancelled
* Implement Unregister helper method
- Implement Unregister helper method, each indicator which hold a
reference to the consolidators updating it. Adding regression
algorithms
* Fix for QuantBook indicators
* Address reviews
* Fix EMA indicator first value calculation
As done by TALib and TradingView, the first EMA value after warmup is
the SMA of the first period.
* Update TSI indicator test data
Data was exported from TradingView
* Update TRIX indicator test data from TradingView
* Update AccumulationDistributionOscillator indicator test data from TradingView
* Update Double EMA indicator test data from TradingView
* Update McClellanSummationIndex indicator test data
* Update SchaffTrendCycle indicator test data
* Update TripleExponentialMovingAverage indicator test data from TradingView
* Update stats for algorithms using EMA
* Update failing unit tests
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Indicator
* Fixed logic error
* Removed old files, added KER into KAMA
* Removed old comments
* Added requested changes
* Minor clean up
* Refactors KaufmanEfficiencyRatio and KaufmanAdaptiveMovingAverage
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
Co-authored-by: Alexandre Catarino <AlexCatarino@users.noreply.github.com>
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Updates KeltnerChannels to Update MiddleBand with EndTime
* Adds Unit Test
This unit test shows that the timestamp of the MiddleBand and the indication are aligned as expected.
Build & Test Lean / build (push) Has been cancelled
Regression Tests / build (push) Has been cancelled
* Super Trend Indicator #4653
* Updated test data
Previous test data was wrong
* Reduced if statement
* Updated tests
Replacing spy test data with dwac test data from trading view.
* Minor comment update
* Minor tweaks
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* First BetaIndicator prototype
- In construction
* Fist BetaIndicator version and unit tests
* More unit tests and regression test
* Nit change
* Requested changes
* Nit changes
* Requested changes
* Adjust beta formula slightly and nit changes
* Nit change
Build & Test Lean / build (push) Has been cancelled
Regression Tests / build (push) Has been cancelled
* Implement IIndicatorWarmUpPeriodProvider
- Implement IIndicatorWarmUpPeriodProvider in PythonIndicator.cs
- Make a unit test to check whether the WarmUpPeriod is working as expected
- Make a regression test to check the new feature at a system level
* Nit change
* Change Period parameter for WarmUpPeriod parameter
- Change regression test to check if the new parameter keep backwards compatibility with indicators that do not set WarmUpPeriod
* Documentation change
* Fix tests bugs
- In CommonIndicatorTests.cs before finish the test it checks the period.value with the number of samples but for default the period.value was set to -1
* Change names
* Change WarmUp and RegisterIndicator methods
- Lean WarmUp indicator skip custom python indicators that don't define WarmUpPeriod parameter
* Call WarmUpIndicator manually
- Add a new "bridge" method called WarmUpIndicator in QCAlgorithm.Python.cs to set up everything to call WarmUpIndicator in QCAlgorithm.Indicators.cs
- Change the regression algorithm to warm up the indicators manually
* Remove unnecessary code and add more tests
* Nit change
* Revert "Nit change"
This reverts commit da411f59c9e4295d75a11c6c581f615a938dd2c6.
* Fix bugs
* Try fix bugs
* Add C# regression test
- More nit changes
- Fix bugs
* Requested changes
* Remove unnecessary code
* Requested changes
* Nit changes
- Add new Python class to check a custom indicator, which doesn't inherits from PythonIndicator, warms up properly
* Reduce redundant code
* Fix bug and add more unit and regression tests
* - Add more unit tests
* Nit change
* Test cleanup
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
Build & Test Lean / build (push) Has been cancelled
Regression Tests / build (push) Has been cancelled
- Remove unnecessary attribute `_period` in MarketProfile
- Set the value of WarmUpPeriod to the period in MarketProfile constructor
- Fix unit test to check VolumeProfile and TimeProfile indicators are being warmed up properly and its WarmUpPeriod parameter isn't zero
* Implement IIndicatorWarmUpPeriodProvider
- Implement IIndicatorWarmUpPeriodProvider in PythonIndicator.cs
- Make a unit test to check whether the WarmUpPeriod is working as expected
- Make a regression test to check the new feature at a system level
* Nit change
* Change Period parameter for WarmUpPeriod parameter
- Change regression test to check if the new parameter keep backwards compatibility with indicators that do not set WarmUpPeriod
* Documentation change
* Fix tests bugs
- In CommonIndicatorTests.cs before finish the test it checks the period.value with the number of samples but for default the period.value was set to -1
* Change names
* Change WarmUp and RegisterIndicator methods
- Lean WarmUp indicator skip custom python indicators that don't define WarmUpPeriod parameter
* Call WarmUpIndicator manually
- Add a new "bridge" method called WarmUpIndicator in QCAlgorithm.Python.cs to set up everything to call WarmUpIndicator in QCAlgorithm.Indicators.cs
- Change the regression algorithm to warm up the indicators manually
* Remove unnecessary code and add more tests
* Nit change
* Revert "Nit change"
This reverts commit da411f59c9e4295d75a11c6c581f615a938dd2c6.
* Fix bugs
* Try fix bugs
* Add C# regression test
- More nit changes
- Fix bugs
* Requested changes
* Remove unnecessary code
* Requested changes
* Nit changes
- Add new Python class to check a custom indicator, which doesn't inherits from PythonIndicator, warms up properly
* Reduce redundant code
Build & Test Lean / build (push) Has been cancelled
Regression Tests / build (push) Has been cancelled
* Add base class for pivotpointshighlow
* abstracts do not work well with python switch to virtuals and use pivotpointhighlow as base implementation
* added missing documentatino header
* moved reset back to original location
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
- Implement IIndicatorWarmUpPeriodProvider interface in both indicators
- Add test to check the new functionality is working as expected. To test WindowIndicator IIndicatorWarmUpPeriodProvider there was used WindowIdentity indicator tests because that indicator only inherits from it
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Refactor for two types in composite indicator
* Fixes for unit tests
* Non generic indicator base
* Further seperate non-generic indicator base and generic input base
* Adjustments
* Remove generic CompositeIndicator
* Add test suite
* Make each method an individual case
* Address review
* Testing cleanup
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Add ValueAreaVolumePercentage Parameter
Remove hard coding of parameter and set it to optional parameter for the user.
* Update TimeProfile.cs to include valueAreaVolumePercentage parameter
* Update TimeProfile definition in QCAlgorithm.Indicators.cs
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Create Market Profile indicator.
* - Requested changes in MarketProfile indicator made it.
- TimeProfile and VolumeProfile created and tested.
- tp_datatest.csv and vp_datatest.csv extracted from https://github.com/bfolkens/py-market-profile
* Final potential Market Profile Indicator.
All unit test made it and the indicator is passing all of them
Test cases made it with python library from GH issue
* Code styling request changes
* Minor suggestions
* Minor renaming tweaks
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
Ricardo Andrés Marino Rojas