2013-12-05 Daniel Sabanes Bove
* move examples directory with simulation to inst directory to
comply with CRAN standards, plus linewrap in man file.
2013-11-22 Daniel Sabanes Bove
* Very minor internal change to avoid possibly invalid coercions
in C++ code (thanks to Brian Ripley for pointing this out).
2012-05-30 Daniel Sabanes Bove
* Minor changes in hope to pass R CMD check on R-Forge.
2012-04-25 Daniel Sabanes Bove
* R/plotCurveEstimate.BmaSamples.R: add "median" as output
* R/BmaSamples.R: add option to include zero samples
2012-03-01 Daniel Sabanes Bove
* src/various: Use Rcpp for creation of results lists, in order to
avoid strange memory-suspicious errors.
2012-02-22 Daniel Sabanes Bove
* src/hyperg.cpp: Corrected the computation of the Laplace
approximation of the logPsi function.
2011-10-07 Daniel Sabanes Bove
* inst/CITATION: The paper is now published in Statistics and
Computing, so I included the vol/number/page info.
2011-01-26 Daniel Sabanes Bove
* R/plotCurveEstimate: Add options "partialResids" and "hpd" to
the two plot functions, to remove partial residuals and to have
equi-tailed instead of highest posterior density intervals /
bands, respectively.
2011-01-18 Daniel Sabanes Bove
* src/bayesMfp.cpp: Changed the logMargLikConst. This is rather
arbitrary if only BFP models are compared, but now it matches the
definition in the hypergsplines package.
2011-01-14 Daniel Sabanes Bove
* Increment version because I will implement a dependent model
prior for the fractional polynomials.
2010-10-29 Daniel Sabanes Bove
* R/BayesMfp.R: add additional attribute "linearInclusionProbs",
which gives for FP terms the posterior probability of exactly
linear inclusion.
2010-09-23 Daniel Sabanes Bove
* src/various: some clarifications for the Solaris C++ compiler
which threw errors on CRAN (e.g. cast integer to double before
taking the logarithm of a number to avoid overloading ambiguity
between float and double casting)
2010-09-20 Daniel Sabanes Bove
* data/ozone.RData: Remove UTF8 degree symbol from Fahrenheit unit
name to avoid a note by R CMD check.
* inst/CITATION: Input the Statistics and Computing paper instead
of the technical report here.
* examples/simulation.R: Add more comments and more sensible
names.
* demo/ozone.R: Add more comments and delete save/load-pairs.
* DESCRIPTION: Add "Hmisc" and "mfp" to suggested packages because
they are loaded in the ozone demo.
2010-06-18 Daniel Sabanes Bove
* R/transformMfp.R (transformMfp): Fix a bug in this
transformation function, which lead to incorrect design matrix
construction when the order of covariates in the mfp object was
different from the order in the BayesMfp object.
2010-06-17 Daniel Sabanes Bove
* Correct prediction methods to use the same shifts of the design
matrix columns as for the original data.
2010-06-16 Daniel Sabanes Bove
* Add a "predictMeans" element to the return value of
"BmaSamples".
2010-03-03 Daniel Sabanes Bove
* Also accept a "main" argument in the plotCurveEstimate functions
2009-12-04 Daniel Sabanes Bove
* Add option to restrict the size of the model cache during model
sampling, to avoid running out of allocatable memory in long model
sampling chains.
2009-09-30 Daniel Sabanes Bove
* Add some documentation to R functions.
* Clearer structure for src directory, so that all newmat stuff is
in a separate directory.
2009-09-29 Daniel Sabanes Bove
* Upload to R-forge SVN.
2009-09-21 Daniel Sabanes Bove
* Only allow identical FP degrees for the covariates, to ensure
that the SWITCH always works.
* Start implementing the new SWITCH move type, which shall guard
us against bad mixing in data sets with near-collinear covariates.
* Note that the R CMD check warning for the marked UTF8 character
is the degree symbol in the labels of the ozone covariates.
2009-09-14 Daniel Sabanes Bove
* start cleaning before the next major revisions
2009-02-12 Daniel Sabanes Bove
* faster implementation for findModel.
2008-12-26 Daniel Sabanes Bove
* test the corrected computation of marginal likelihood (see
hyperg.cpp).
2008-11-29 Daniel Sabanes Bove
* Added flat model prior option as the default prior specification.
2008-10-24 Daniel Sabanes Bove
* first version which is tested for the ozone data.
2008-08-22 Daniel Sabanes Bove
* copied everything from original BayesMfp package. This shall be
a seperate implementation for the new methodology with
hyper-g-prior & Co.