dlmtree 1.2.0
- added support for cluster random effects
- access by using the ‘cluster’ input into any modeling function
- updated sampling methods in HDLM shared to sample entire trees
(vs. grow/prune/change)
- added logit model support for HDLM shared
bug fixes: * issue computing cumulative effects in estDLM.R
dlmtree 1.1.1
- removed C++11 from DESCRIPTION per CRAN requirements
dlmtree 1.1.0
- new feature:
diagnose() function - launches a Shiny
panel for assessing MCMC convergence, which includes:
- trace plots and density plots for the estimated distributed lag
function
- lag-wise convergence analysis
- tree update convergence
- trace plots and density plots for hyperparameters
- exposure selection (for mixture models)
- modifier selection (for heterogeneous models)
- update
summary.X() functions
- more consistent output formats
- add options to extract MCMC posterior samples
- documentation improvement
- add more detailed descriptions in
Value sections
- reduced redundancy across help files of S3method
dlmtree 1.0.1
- change the documentation organization format of the main function
using
list() for input argument
- lists of arguments are linked to
control.R
- remove the prefixes from control arguments
- remove following scripts from the package:
tdlmm.R,
tdlnm.R
- update credible interval bands to be transparent
dlmtree 1.0.0.2
- fix to
shiny.hdlm and shiny.hdlmm to
remove bug where no data is selected
- fix to
combine.models.R to correctly combine tree
posterior samples
- fix to
summary.tdlmm in rowMeans to stop autodrop of
dimension
dlmtree 1.0.0.1
- a split point for Subgroup tab in shiny interface set to the mode of
proportion
dlmtree 1.0.0
- initial CRAN release
- spelling checks using
spell_check_package() function
from spelling package
- S3 method
predict.dlmtree is separately defined as
predict.hdlm and predict.hdlmm
- combined model fitting functions
tdlnm and
tdlmm into dlmtree
dlmtree 0.9.0
- Beta prior for monotone CW selection
- Bayes factor selection method
- zero inflated tree method
- monotone DLNM
- precalculation in tdlnm for large speed improvement
dlmtree 0.8.1.1
- fix to single covariate model
- add functions for posterior analysis with dlmtree class
(
pip and splitpoints)
dlmtree 0.8.1.0
- fix to allow intercept only models
- updated tdlnm algorithm for increased performance (memory
restriction for large datasets)
dlmtree 0.8.0.0
- add HDLM functionality (
dlmtree)
dlmtree 0.7.1.1
- fix
summary.tdlmm if nMix=1 or nExp=1
dlmtree 0.7.1.0
- combined identity and logistic link into single c++ file
- several code efficiencies
- updated print output of
tdlmm.summary
- removed prior on dirichlet hyperparameter
- exposure selection based on Beta-Binomial distribution
keep.mcmc in summary
dlmtree 0.6.2.0
- separated
tdlmm from tdlnm function
- included exposure selection into summary
- prior on dirichlet hyperparameter ~Gamma(1, nTrees/4)