Skip to content

Releases: JacobSeedorff21/BranchGLM

Version 2.1.6

12 Jun 00:15
Compare
Choose a tag to compare

Adding a basic plot method for BranchGLM objects and fixing a bug in the calculation of the log-likelihood for gaussian regression models with an odd number of observations.

Version 2.1.5

08 Apr 02:02
Compare
Choose a tag to compare

Fixing a bug in parallel computation for the branch and bound algorithms. Also updated stepwise variable selection algorithms to return the whole search path and deprecated maxsize argument

Version 2.1.4

27 Jan 18:40
Compare
Choose a tag to compare

Updating all documentation, improving tests, fixing some bugs, and improving consistency between fitting methods

Version 2.1.3

06 Dec 21:55
Compare
Choose a tag to compare

Adding cols argument to plot.BranchGLMVS to control the colors used, improving predict and coef functions, and deprecating fit.

Version 2.1.2

31 Aug 00:48
Compare
Choose a tag to compare

Adding some new arguments to plot.BranchGLMVS and fixing a bug in the same function. Also fixing a bug in the confint.BranchGLM function

Version 2.1.1

14 May 16:52
Compare
Choose a tag to compare

Adding new arguments to VariableSelection, plotCI and plot.BranchGLMVS functions.

Version 2.1.0

04 Mar 18:11
Compare
Choose a tag to compare

Adding confidence intervals for regression models and fixing bugs.

Version 2.0.1

22 Nov 22:47
Compare
Choose a tag to compare

Fixing bugs in VariableSelection function and adding addLines argument to plot.summaryBranchGLMVS function.

Version 2.0.0

31 Oct 20:34
Compare
Choose a tag to compare

The VariableSelection function can now find the best k models according to the desired metric. Additional functions were added to help work with and visualize the results from VariableSelection when finding multiple best models.

Version 1.3.2

04 Oct 00:34
Compare
Choose a tag to compare
Version 1.3.2 Pre-release
Pre-release

Slightly speeding up variable selection with branch and bound methods. Also fixing bug when using gamma regression when performing variable selection.