Changelog¶

1.5.0 2022-01-24

[Feature] #83: Added Sparse Sequential Bootstrap method to Sampling Module (up to 200 times faster execution).
[Feature] #86: Speed Up the Fractionally Differentiated Features from the Feature Engineering Module (up to 10 times faster execution).
[Feature] #88: Improved the visualization of the Model Fingerprint Algorithm.
[Feature] #91: Noise Reduction methods - KCA, FFT, and LOWESS added to the Feature Engineering Module.
[Feature] #92: Directional Change added to the Feature Engineering Module.
[Feature] #93: Volatility Estimators added to the Feature Engineering Module.
[Feature] #94: Entropy Measures added to the Microstructural Features Module.
[Feature] #94: First Generation Microstructural Features added to the Microstructural Features Module.
[Support] #83: Updated Sequential Bootstrap and Sequentially Bootstrapped Ensembles documentation.
[Support] #85: Added presentation slides and videos to documentation.
[Support] #87: Added missing notebooks to documentation.
[Support] #89: Added blog post links to documentation.
[Support] #90: Added Futures Roll method to documentation.
[Support] #91: Noise Reduction methods - KCA, FFT, and LOWESS documentation.
[Support] #92: Directional Change documentation.
[Support] #93: Volatility Estimators documentation.
[Support] #94: Entropy Measures documentation.
[Support] #94: First Generation Microstructural Features documentation.

[Feature] #72: Changed the analytics we track to: MAC Address, Public IP, API_KEY, and Function Calls + time stamps.
[Bug] #72: Fixed the API key validation and build server authentication.
[Bug] #72: Fixed maxed connections a day by pointing the get public IP server to AWS.
[Bug] #77: Fixed modification Jacques made to 3B that used log returns instead of returns.
[Bug] #77: Deprecated the stacked sample weights.
[Bug] #77: Fixed broken builds due to number of models used in seq boot ensemble.
[Support] #75: Docs: Improved documentation for user experience.

[Feature] #69: Requirements versions are now non-fixed.
[Feature] #69: Added support for Python 3.6 and Python 3.7.
[Support] #69: Reflected Optimal Mean Reversion Module migration in the documentation.
[Support] #69: Migrated Optimal Mean Reversion Module from MlFinLab to ArbitrageLab.

[Feature] #46: Lambda code in Microstructural Features Module speed-up.
[Feature] #46: Stacked Module with Cross Validation, Feature Importance, and Sampling methods added.
[Feature] #45: Added Pagan et al. and Lunde et al. Bull Bear Methods to the Labeling Module.
[Feature] #59: Code and unit tests style unified.
[Feature] #61: History Weighted Regression added to the Regression Module.
[Feature] #58: Low silhouette scores check made optional in Feature Clusters Module.
[Feature] #56: MAE/MSE added as possible metrics for the Trend Scanning Module.
[Bug] #60: Fix structural break bug in the Chu-Stinchcombe-White test.
[Bug] #57: Fix purging bug in Purged KFold/Combinatorial Purged KFold.
[Support] #55: Removed TensorFlow from requirements and adjusted installation guide.
[Support] #46: Stacked Module documentation.
[Support] #45: Added Pagan et al. and Lunde et al. Bull Bear Methods documentation.
[Support] #59: Documentation style unified.
[Support] #61: History Weighted Regression documentation.

[Feature] #52: Migrated Online Portfolio Selection Module code from MlFinLab to PortfolioLab.
[Feature] #52: Migrated Portfolio Optimisation Module code from MlFinLab to PortfolioLab.
[Feature] #50: Added t-student option to BVC classifier.
[Feature] #44: Added n_repeat parameter to MDA feature importance.
[Feature] #35: Debugged ETF Trick code.
[Bug] #50: Fix bug in Bar-based Kyle lambdas calculation.
[Support] #52: Updated requirements versions (numpy==1.20.1, matplotlib==3.2.2, pandas==1.1.5, scikit-learn==0.24.1, scipy==1.6.0, statsmodels==0.12.2).
[Support] #52: Migrated Online Portfolio Selection Module documentation from MlFinLab to PortfolioLab.
[Support] #52: Migrated Portfolio Optimisation Module documentation from MlFinLab to PortfolioLab.