- Fixed invalid email address for Yakuan Chen
- Fix threshold for small sigma2 in mfpca.face function to ensure accurate score estimation for level1 scores
- Added parameters npc2 and pve2 to mfpca.face to allow for user to specify separate npc or pve for level 2 decomposition
- One line fix to pfr that allows pfr to be called from within another function
- Pull request on URL updates in documentation
- Added pve to what is returned by mfpca.face
- Change names in list of what is returned by mfpca.face to be the same as mfpca.sc
- Updated email address for Erjia Cui
- Changed Erjia from contributor to author
- Added pve to what is returned by fpca.sc and fpca.face
- Simon Wood removed "pers" argument from mgcv::plot.random.effect. Removed reference to pers from plot_pfr.gam to avoid documentation and code being inconsistent.
- added COVID19 datasets
- minor bug fix for mfpca.face
- updated email address for package maintainer
- removed import lattice::qq from pffr-methods.R
- added periodic spline option for
fpca.face - changed if(class(object) != "string") to if(!inherits(object, "string")) in
ccb.fpc.Randfosr.perm.test.Rfiles to fix Note.
- bug fix for
pfrmodels without intercept (thx, @ZheyuanLi)
- New function,
mfpca.face, which is a faster version ofmfpca.sc
- Minor bug fixes in
fpca.faceto patch error when npc = 1
- Minor bug fixes in
pfrto patch error in R version 4.1 - Updated documentation URLs
- Minor bug fixes in
fgamexamples for upcoming R release
- Fixes bugs
- Commented out option
useSymm = TRUEin tests forfpca.sc
- Commented out option
- Fixes a bug in
fpc()due to release of R 4.0.0 that changes the following:
R> class(matrix(1 : 4, 2, 2))
[1] "matrix" "array"
(and no longer just "matrix" as before), and that conditions of length
greater than one in 'if' and 'while' statements executing in the package
being checked give an error.
- change email address for maintainer
-
fix minor bug in rlrt.pfr.R
-
change maintainer from Rayman Huang to Julia Wrobel
-
updates for compatibility with mgcv 1.8-23 (#69 etc.)
-
fixed fpcr for scalar covariates (#76)
-
now re-exports cmdscale_lanczos
-
homogenized inputs/outputs to most fpca.XXX functions
-
fix documentation for fpca.face
-
added pco ridge regression see ?poridge
-
add
fpca.lfda()function -
add functions and data set from refund.shiny package
-
add
mfpca.sc()function. -
add example for DTI data.
-
export
pfr_old(). -
fix documentation and add warning message for
rlrt.pfr().
-
new
pfr()function. The newpfr()function merged the old pfr and fgam functionality. -
Switch to roxygen2 documentation
-
allow an "argvals" argument to all relevant functions for consistent strucuture throughout the package
-
refactor input and output argument for
fosr(). -
add vignettes.
-
bug fix in
fpca.face(). -
export
expand.call(). -
add cross-sectional FoSR using GLS, variational Bayes and Gibbs sampler