NEWS.md

Version 0.1.8, 2016-07-01

FEATURES / CHANGES

• coef.glmnetLRC() receieves a tolerance argument to avoid selecting coefficients that are very close to 0.

Version 0.1.7, 2016-06-21

FIXES

• predict.glmnetLRC() now correctly handles matrices as predictors, as well as data frames.

Version 0.1.6, 2016-06-17

FEATURES / CHANGES

• Removing use of Smisc::sortDF() in single_glmnetLRC()

Version 0.1.5, 2016-05-23

FEATURES / CHANGES

• Added plot.LRCpred() to plot the predicted probabilites of the LRC, along with associated tests
• In summary.LRCpred(): Included summary of predicted probabilities, added print.summaryLRCpred()
• Tightened up the argument checking of predLoss_glmnetLRC()

Version 0.1.4, 2016-05-11

FEATURES / CHANGES

• Added prediction probabilities to the output of the predict.glmnetLRC() method

Version 0.1.3, 2016-05-04

FIXES

• Corrected the way predict.glmnetLRC() handled additional columns specified by keepCols

FEATURES / CHANGES

• Added the missingpreds methods to easily identify needed predictors that may not be present in new data
• Remove uneeded documentation for generic method, extract

Version 0.1.2, 2016-04-13

FEATURES / CHANGES

• Package vignette moved to online docs
• Minor documentation edits

Version 0.1.1, 2016-03-09

FEATURES / CHANGES

• Addition of tests against manual fitting with standardized predictors

Version 0.1.0, 2016-02-15

FEATURES / CHANGES

• Original package deployment to github