BAS 2.0.2

Bug Fixes

• minor patch to address rchk warnings about unprotected variables when calling allocating functions.
  Added PROTECT samplemarg in model_probabilities.c

BAS 2.0.0

Features

• in function bas.lm() and bas.glm() replace use of non-API call to SETLENGTH for MCMC sampling with
  new C function resizeVectors to truncate when overallocated (closes issue #82). This should reduce memory allocation
  for large problems when n.models was much larger than needed, as the longer vectors are available for
  garbage collection, although there may be a slight increase in memory usage while copying the to the new vector. This
  uses the C function resizeVectors based on the R internal function xlenghtgets, but uses memcpy to copy the
  data to the new vector for efficiency in the case of real and integer SEXP vectors.

• in bas.lm() and bas.glm, added logical option GROW" for method = "MCMC", to allow output vectors to grow as needed (closes issue #91) rather than over-allocating based on n.modelsand truncating. This should reduce memory usage when the number of unique models visited is much smaller than the default forn.models. Additional options expandandn.models.initcontrol the growth rate and initial size of the output vectors. By default,GROW = TRUE, expand = 1.25andn.models.init = 2500`. See documentation for more details.

• in bas.lm and bas.glm, MCMC sampling now stops after MCMC.iterations, even if n.models is reached,
  improving estimation based on MCMC frequencies.

Bug Fixes

• R2 is calculated correctly in models where the number of columns in the design matrix was greater than
  n, but the model was full rank. Closes issue #96 reported by A. Womack.

• Fixes issues #97 reported by A. Womack where the truncated Poisson and truncated power prior probabilities
  did not account for the number of models of a given size.

BAS 1.7.5

Version 1.7.5 of the BAS package adds an internal function to count the number of models
that satisfy "hereditary" constraints. This is used in the force.heredity option
in bas.lm to reduce the number of models considered in the sampling process and should
reduce the memory requirements and speed up the sampling process. This currently works only
for factors included in the model formula, but not with factors always included in the model
orwith other hereditaty constraints such as with polynomials. (theforce.heredity option
does work with these other constraints). This is a first step in reducing the number of models
allocated in the sampling process. Future updates will include other hereditary constraints.

BAS 1.7.3

BAS 1.7.3 introduces a new Independent Adaptive MCMC algorithm for bas.lm that can be used a proposal distribution for sampling models with replacement and estimation of posterior model probabilities via Importance sampling and Horvitz-Thomposon estimators and their Bayesian Finite Population estimators. See details in bas.lm with `method="AMCMC".

BAS_1.7.3.tar.gz

BAS 1.7.2

Updated package provides a new adaptive independent MCMC sampler that allows more accurate estimates of model probabilities and other quantities using the Horvitz-Thompson estimator and Bayesian analogs for finite population sampling

Full Changelog: v1.7.1...v1.7.2

BAS 1.7.1

Minor Improvements and Fixes

• Initialized vector se via memset and disp = 1.0 in fit_glm.c (issue #72)

• Initialized variables in hyp1f1.c from testthat (issue #75)

• Removed models that have zero prior probability in bas.lm and bas.glm (issue #74)

• Fixed error in bayesglm.fit to check arguments x or y for correct type before calling C and added unit test (issue #67)

BAS 1.6.6

New Features

• Added support for Gamma regression for bas.glm, with unit tests and
  example (Code contributed by @betsyberrson)

Minor Improvements and Fixes

• added error if supplied initial model for the bas.lm sampling methods "MCMC" and "MCMC+BAS" had prior probability zero.

• fixed printing problems as identified via checks

• fixed indexing error for bas.lm and method = "MCMC+BAS" as bas.lm using method = "MCMC+BAS" crashed with a segmentation fault if bestmodel is not NULL or the null model. GitHub issue #69

• fixed error in predict.bas with se.fit=TRUE if there is only one predictor. GitHub issue #68 reported by @AleCarminati
  added unit test to test-predict.R

• Fixed error in coef for bas.glm objects when using a betaprior of class
  IC, including AIC and BIC Github issue #65

• Fixed error when using Jeffreys prior in bas.glm with the
  include.always option and added unit test in test-bas-glm.R.
  Github issue #61

• Fixed error for extracting coefficients from the median probability model
  when a formula is passed as an object rather than a literal, and added
  a unit test to test-coefficients.R Github issues #39 and #56

v1.6.4

Latest release for CRAN

BAS version 1.6.2

Release for updates with R 4.2.0

Major change is improved behavior for CCH and related priors in bas.glm that use the phi1 function. Alternative formulations for computing the marginal likelihoods show add improved stability and eliminate/reduce NA and Inf in computations as reported in Issue #55

BAS version 1.6.0

Changes

• update Fortran code to be compliant with USE_FC_LEN_T for character strings

Bug Fixes

• fixed warning in src code for log_laplace_F21 which had an uninitialized variable
  leading to NaN being returned from R function hypergeometric2F1