chainladder-python

patch to 0.6.1

Bug fixes:

• Corrected a bug where TailConstant couldn't decay when the contant is set to 1.0
• #71 Fixed issue where ``Pipeline.predictwould not honor thesample_weight` argument

Enhancements:

• #72 Added drop method to Triangle similar to pd.DataFrame.drop for dropping columns
• Added xlcompose yaml templating
• #74 Dropped link ratios now show as ommitted when callinng link_ratio on a Development transformed triangle
• #73 Triangle.grain now has a trailing argument that will aggregate triangle on a trailing basis

Enhancements

• Added TailBondy method
• Propagate std_err_ and sigma_ on determinsitic tails in line with Mack for better compatibility with MackChainladder
• Improved consistency between to_frame and __repr__ for 2D triangles.

Bug Fixes

• Fixed a bug where the latest origin period was dropped from Triangle initialization when sure data was present
• resolves #69 where datetime was being mishandled when ingested into Triangle.

Bug fixes

• resolves #62 Slicing cdf not working correctly
• resolves #61 groupby not consistent with pandas
• resolves #60 Development.drop_high and drop_low. drop_high and drop_low only drop one observation even if multiple observations match the criteria.
• resolves a bug in TailCurve where the fit_period slicer was not indexing properly
• Better support for the loc property of a Triangle.

Enhancements:

• Resolves #64 support pandas>=1.0.0. chainladder now works with pandas>=1.0.0
• New feature #63 fill values for undefined link ratio. fillna is a new parameter of the Development estimator
• New feature #58 Attachment age support in Tail methods. attachment_age is a new parameter of the TailCurve estimator.
• New feature #59 Add cl.concat functionality
• Added ability to perform arithmetic between a Triangle and an np.ndarray

Enhancements

• Resolves #31 Propagate bootstrap process risk. This should make the BootstrapODPSample estimator fully functional.
• Enhanced Benktander and BornhuetterFerguson methods to allow for stochastic apriori picks.

Bug fixes

• MackChainladder was incorrectly calculating MSE when dropping link ratios
• Fixed issue where using grain always returned a development triangle even when fed a valuation triangle

Enhancements

• Added __round__ and __pow__ dunder methods
• Warn on triangles with invalid origin/development combinations

Bug fix

• patch to column slicers.

Enhancements:

• Substantial speed up in triangle initialization for larger triangles
• Fancy loc/iloc indexing now honors the order the user chooses

Bug fixes:

• Mack Std Err for tail fix xonsistent with R v0.2.11 fix
• Valuation triangles should now be JSON serializable
• Bug in expected loss method predict that wasn't rendering prediction results correctly

Enhancements:

• Full GPU-support using cupy.

Bugfix

• quick patch for Pipeline

Bug fixes:

• Fixed an inappropriate mutation in the Benktander estimator when calls to predict are made
• Allow for sample_weight argument to be directly passed to a Pipeline

Enhancements:

• Added broadcast_axis for better BootstrapODPSample support

Bug fixes:

• Eliminated a side-effect in Development when using drop
• Improved dropping capabilities of BootstrapODPSample

Bug Fixes:

• Fixed TailCurve shape misaligned #57
• Fixed triangle json serializers to include all data necessary to recreate triangle
• Fixed various edge case issues when len(development) == 1
• Fixed MackChainladder assymentric triangle support (e.g. OYDQ)
• Fixed MunichAdjustment assymentric triangle support (e.g. OYDQ)
• Fixed datetime flexibility on dataframe ingestion into triangle

Enhancements:

• aligned append method with pandas

Other:

• Moved excel exhibit functionality to a separate xlcomose package.

Bug fixes:

• patched a defect in Triangle.grain()
• patched a defect in exhibits.py pertaining to titles

Bug Fix:

• Fixed issues in asymmetric triangles

Enhancement:

• Replaced Exhibit API with Row, Column, Tabs, and DataFrame

Bug Fixes:

Minor release to improved commutative properties of triangle functions (grain, val_to_dev, dev_to_val, cum_to_incr, incr_to_cum, latest_diagonal) for all types of triangles (development triangles, valuation triangles, complerte triangles)

Bug fix

• patch to asymmetric LDF fix in release 0.4.0

Enhancement

• Added Triangle.dropna() method that allows shaving off origins/development vectors that are all 0/NA

Enhancements:

• Added better datetime management when instantiating a triangle
• Added json serializers for Triangle as well as the estimators and Pipeline. These can be accessed with the to_json method and cl.read_json
• Added DevelopmentConstant estimator that takes LDF or CDF patterns as a dictionary instead of calculating from a triangle. This is useful for incorporating industry patterns into an analysis
• General improved functionality for malformed triangles
• improved memory management by eliminating unnecessary usage of deepcopy

Bug Fixes:

• Fixed an issue where LDFs were calculated incorrectly for asymmetric triangles
• Fixed a bug where slicers weren't working properly after particular mutations (e.g. adding a new column)
• Arithmetic on triangles with different origin ranges will now take the union of the origin periods. Previously the intersection was taken.