Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Release v2.2.0 #65

Merged
merged 11 commits into from
Sep 3, 2019
Merged

Release v2.2.0 #65

merged 11 commits into from
Sep 3, 2019

Conversation

juliangehring
Copy link
Owner

Version 2.2.0

Changes

kleinschmidt and others added 11 commits July 27, 2019 19:56
This is a different approach to upgrading support for StatsModels to 0.6. It
takes advantage of Terms 2.0 (Son of Terms) functionality by creating a special
Term type that wraps the left-hand side of the formula, and returns a sampled
version of the response variable every time it's asked to generate model
columns. This allows the two separate methods for bootstrap using different
sampling strategies to be combined by factoring out the response re-sampling
into a single place. More importantly, making the resampling step lazy in this
way removes all of the bookkeeping and copying of data, since the residuals are
resampled and added to the response at the last moment.

It also allows for any of the terms types supported by terms 2.0 to be used on
the LHS, including arbitrary functions. So you can do things like
@formula(log10(X) ~ log10(U)) and it will work (and the tests have been updated
to reflect this)

Includes:
* bump statsmodels to 0.6 and Formula->FormulaTerm
* guard against FunctionTerm on LHS of formula
* use ResidualTerm for resampling
* use transformed formulae in parameteric bootstrap tests
* Format code and clean comments
* Drop explicit precompile statment
* Fix package version number
* Stop exporting `iquantile`
* Update travis configuration
Drop manifest and test require files
Fix deprecations in `DataFrames` and `GLM`
@juliangehring juliangehring merged commit a447c29 into master Sep 3, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants