| NEWS | R Documentation |
suppress some warnings in tests for cross-platform/CRAN compatibility
intercept-only parameters were handled wrong (GH #8)
fix error-handling (class vs. inherits) for R 4.0 compatibility
pop_pred_samp to compute population prediction
samples, tweaking non-positive-definite covariance matrices and using
importance samples if necessary/specified. This is
still VERY experimental/in flux!
added dnorm_n function for Normal densities
with the standard deviation parameter profiled out
add slnorm function for log-Normal densities
(median and mean only for now)
IC tables now have an as.data.frame method (from
@iago-pssjd on GitHub)
the evaluated
start and fixed components of the call are now stored
in the call component (Daniel B. Stouffer)
fixed buglet: flipped profile plot axes, confint for negative values
summary.mle2 is now exported, for use by other
packages
small fixes to AIC(c) methods
fixed bug: evaluate call$method so that
profiling/updating works within a function environment
make AICtab smarter about NA values
fix BIC bug (infinite recursion)
hessian computation uses gradient function if provided
basic continuation method implemented for profiling (with stubs for smarter methods)
mle2 stores its calling environment for more flexibility when re-evaluating, e.g. in profiling (could lead to occasional surprises, e.g. if saving a fitted mle2 object with large objects in its calling environment)
update slice functionality; allow for explicit ranges
CRAN updates (import from base packages)
new warning if ~dnorm is called with sd implicitly ==1
some internal tweaking to slice functionality
updated to allow for MuMIn v 1.10.0 changes
fix mnames behaviour
slight vignette cleanup
add logLik option to IC tables, more use of nobs methods
minor improvements to slice functionality
more CRAN tweaks
.Rbuildignore tweaks for CRAN
vignette tweaks
add Depends: R>=3.0.0
add .Rbuildignore for CRAN tests
adapt to updated optimx
tweaks for CRAN compliance
switch from aod to aods3 in vignettes
switch vignette to knitr
tweaks to print.ICtab()
warn on convergence failure
fixed (fairly obscure) bug in assigning data environment to gradient function; replace 'gr' in call with appropriately evaluated version
changed some cat() to message()
=======
add .Rbuildignore for CRAN tests
adapt to updated optimx
tweaks for CRAN compliance
switch from aod to aods3 in vignettes
switch vignette to knitr
tweaks to print.ICtab()
warn on convergence failure
fixed (fairly obscure) bug in assigning data environment to gradient function; replace 'gr' in call with appropriately evaluated version
changed some cat() to message()
>>>>>>> .r116
remove BIC definitions (now unnecessary/should be adequately defined in core R)
add explicit Depends: on stats4
note that development optimx (2012.05.24+) now fails on profiling when reduced model is 1-dimensional (i.e. for a 2-parameter model)
remove spurious .RData file; version bump
wrapped eigen() call to prevent failure of eigenvalue ratio summary when hessian is bad
fix bug: forgot to export summary method for mle2
add exclude.fixed argument to coef method
fix bug: single lower/upper parameter, or prof.lower/prof.upper parameter, not interpreted properly in profile/confint
add and document slice methods: change from old
(profile-like) behavior, old version is available as sliceOld
DESCRIPTION/NAMESPACE fixes, move most Depends: to Imports: instead (except methods package)
fix bug in gradient/vecpar/profiling interaction (Daniel Kaschek)
improve (and document) uniroot method for confint: now respects box constraints
fix issue where bobyqa (optimx) strips parameter names from result
remove a bit of installed junk; vignette fix
fix imports for latest version of R-devel
modified starting value code and documentation slightly: now allows per-parameter lists. Updated docs.
Fixed bug that would screw things up if 'i' were used as a parameter.
Update vignette to conform to MuMIn changes
fixed buglet in detection of no-intercept models when setting starting values
Turned off some warnings when skip.hessian=TRUE (e.g. in profile)
Calculate max grad and Hessian eigenval ratio, add to "details"
(Should add accessor methods)
Fixed NAMESPACE to export methods etc etc properly; removed universal exportPattern directive
mainly just a version bump for CRAN
added tag for variables in profile/confint warnings
lots more technical information in vignette about profiling algorithm
changed NEWS to NEWS.Rd
fixed bug for "optimize" – profile etc. now respect bounds
eliminated warning about bounds names if all identical
add "try_harder" flag to profiling (ignore flat spots, NAs ...)
gradient functions work better with fixed parameters, hence with profiling
profile plot reverts to linear for non-monotonic profile
added warning in confint for non-monotonic profile, or for non-monotonic spline fit to monotonic profile; revert from spline+linear to linear approximation in this case
various documentation improvements
optimx improvements
require data= argument when using formula interface
turn off hessian computation in profile
allow use of MASS::ginv
bug fix in calc_mle2_function for no-intercept models (thanks to Colin Kremer)
fixed optimx, added 'user' option
changed hessian calculation to use numDeriv code (causes tiny changes to hessian results that could matter in edge cases). Too lazy to provide a backward compatibility mode ...
documented optimizer= choices in ?mle2
fixed bug in AICc (David Harris)
added NAMESPACE, various fixes to go with that
beginnings of an RUnit testing framework
tweaked vignette
added prof.lower, prof.upper to profile()
added "optimize" to list of allowed optimizers, some bug fixes
tweaked par() resetting in profile plots
more qAICc fixing
tweaked handling of bounds: profile now succeeds on some 1D problems where it didn't before
added deviance, residuals methods
added newparams argument to predict, simulate; newdata argument to simulate
added vignette (stub)
added explicit params argument, to help sort out full parameter specifications when parameters is non-NULL
fixed predict() for case with parameters
added snorm
changed ICtab defaults to weight=TRUE, base=FALSE, sort=TRUE
added simulate method (formula interface only)
fix AICctab bug
remove spurious cat/print in profile
fix qAIC bug
fix Tom Hobbs bug: named lower/upper/parscale/ndeps get rearranged properly, otherwise rearrange in order of "start" and issue a warning
documentation tweak for S4 as.data.frame
added sbeta to list of known distributions
removed nlme requirement & auto-loading
version bump, submit to CRAN
added predict method
added flexibility for profile plotting (main, x labels etc.); added examples
added an instance of "namedrop" to fix naming problem
added tol.newmin to slice etc.
added check for numeric return from profile within confint
fixed bugs in profile plotting when profile is restricted to a subset of variables
added tests for par() to reset to original on exit
improved profile documentation
replicate std.err if specified in profile
add as.data.frame
tweak tol.newmin (better fit found during profile) code
version bump, moved to R-forge.
reordered NEWS file (most recent first)
tweaked stop-on-better-fit code
fixed (?) qAIC(c) methods
tweak/fix to ICtab documentation (thanks to Tom Hobbs)
added qAIC(c) methods (not working yet!)
oops. Fixed infelicity (bug?) in new environment manipulation
tweaked environment/data assignment to preserve original minuslogl environment better
changed plot.profile.mle2 options (added onepage etc., made plot.confstr=TRUE by default)
added warning about too-short lower/upper
added documentation
fixed bug in AICctab
cosmetic change to printing – save call.orig
moved ChangeLog to NEWS
fixed (?) environment bug
tried to use built-in relist, but failed: renamed relist to "relist2" (try again later)
documented get.mnames (auxiliary function for ICtabs)
started to add gr (gradient) capability – NOT TESTED
changed ICtab to allow either ICtab(x,y,z) or ICtab(list(x,y,z)) (L <- list(...); if is.list(L[[1]]) && length(L)==1)
fix bug in profiling: all optim() methods EXCEPT L-BFGS-B. return the value of the objective function if given a function with no arguments/zero-length starting parameter vector (this is the situation with "profiling" a 1-D function). L-BFGS-B gives funky answers. added a check for this case. (may need to check behavior for alternate optimizers (nlm etc)) [this behavior triggered a "found better fit" error when profiling 1D functions with L-BFGS-B]
changed behavior when finding better fit during profiling to return new parameters
tweak vignette
fixed second major AICc bug (was fixed in mle2 method, but not in logLik method)
change "ll" to "LL" in examples for clarity
tweaked anova reporting of models (wrap instead of truncating)
added (undocumented) show.points option to profile plot to display actual locations of profile evaluation
tweaked profile to behave better when profiling variables with constraints (upper, lower)
moved vignette to inst/doc where it belongs
ICtab hack to protect against package:aod definition of AIC(logLik)
added submit stub
tweaked slice.mle2-class docs for consistency
fiddled with vignette
preliminary code to allow non-monotonic profiles
preliminary add nlm to list of optimizers (untested)
add aod, Hmisc, emdbook to VignetteDepends and Suggests:
better df extraction in ICtab
minor bug fix for AICc (allows AICc of nls objects)
handle models with -1 in formula better: starting values set "all equal"
made ANOVA formula line-length accessible
added skip.hessian and trace arguments to mle2
messed around with BIC definition – attempt at consistency with nlme
added rudimentary support for nlminb, constrOptim
nlme now required for fdHess (which is required for nlminb since it doesn't compute a finite-diff Hessian)
add experimental formula interface
change all names from mle to mle2 to avoid confusion/conflicts
with stats4 version of mle
change internal structure of data evaluation
worked on vignette
added optimizer slot (stub)
fix AICc bug! (was deviance+2*k*(k+1)/(n-k-1), not AIC+2*k*(k+1)/(n-k-1))
change AIC to AICc for corrections
add AICtab for weights, delta, sort ... options
expose error messages occuring within profile()
uniroot tries harder to find a valid endpoint
truncate terms in anova.mle at 80 characters
enhanced anova method, works with print.anova
tweaked namedrop() code – ??
added parnames, parnames<-
minor fix to allow "profiles" of 1-parameter models (skip fdHess call)
minor change to print method for mle results
tweaking "vecpar" (to allow parameter vectors in objective function)
removed fdHess/nlme dependency