Version 3.15.0

• New: boot_ncvreg() function to obtain confidence intervals
• New: assign_fold() function to assign folds for CV
• Change: seed is no longer an argument to CV functions; use assign_fold() instead

Version 3.14.2

• Documentation: Lots of formatting fixes to the documentation
• Fixed: cv.ncvreg(), cv.ncvsurv() no longer affect seed in global
  environment if seed is specified

Version 3.14

ncvreg 3.14.0 (2023-03-28)

• New: residuals() method
• New: std() can now be applied to new data
• New: summary.ncvreg() now offers sort option; fixes #13
• Change: fir() deprecated
• Change: local_mfdr() allows user to specify sigma; also uses CV if
  called with cv object
• Fixed: Manual color palettes now recycled correctly; fixes #40;
  thank you to Logan Harris for pointing this out
• Fixed: mfdr now works for Poisson
• Documentation: Adding vignettes on other CV criteria, adaptive
  rescaling
• Documentation: References reformatted, URLs updated, DOIs added
• Internal: C code for binomial, poisson now unified under glm
  structure
• Internal: Now using roxygen2 for all documentation

Version 3.13.0

• New: Options xtx and r for ncvfit()
• Fixed: AUC() now compatible with survival 3.2.10
• Internal: cv.ncvreg() now uses less memory (returnX off)
• Internal: Better error handling if a matrix is supplied for y

Version 3.12.0

• New: ncvfit(), a raw API to the ncvreg solver with full control over standardization, etc.
• Changed: ncvreg() and ncvsurv() now issue warning for non-pathwise usage
• Internal: Now using tinytest for unit testing
• Fixed: Memory leak in cox-dh; resolves #20

Version 3.11.2

• New: std() now works on integer matrices and numeric vectors
• Internal: Lots of internal changes for cleaner, more reliable code
• New version numbering system

Version 3.11-1

• Fixed: Leave-one-out cross-validation now works correctly for logistic regression
• Documentation: Added documentation (online) for local mfdr
• Documentation: Fixed some broken links and typos

Version 3.11-0

• Change: returnX now turned on by default if X < 100 Mb (used to be 10 Mb)
• Change: summary.ncvreg now based solely on local mfdr
• Change: Loss functions now consistently defined as deviance for all types of models
• Change: R² now consistently uses the Cox-Snell definition for all types of models
• Change: cv.ncvreg and cv.ncvsurv now return fold assignments
• Fixed: Can now pass fold assignments to cv.ncvsurv
• Documentation: Lots of updates
• Documentation: vignette now html (used to be pdf)
• Documentation: pkgdown website

Version 3.10-0

• New: summary.ncvreg and summary.ncvsurv now report tables of inference for
  each feature based on local mFDRs
• New: Option to specify fold assignments in cv.ncvsurv
• New: CVSE now calculated for Cox models, with option of quick or bootstrap
• Change: returnX now turned on by default if X < 10 Mb
• Change: cv.ncvsurv now balances censoring across fold assignments
• Change: All data sets now follow Data$X, Data$y convention
• Deprecated: cv.ind argument to cv.ncvreg is now called fold
• Portability: Fixed C99 flag
• Internal: Fixed & v && C issue

Version 3.9-1

• Change: Poission now returns linear predictors, like other families
• Internal: Changing PROTECT/UNPROTECT to conform to new coding standards