bamp 3.0.0

Major release. The Polya-Gamma sampler and its native C engine, DIC-based selectModel(), and the identifiability-aware convergence diagnostics in this release were contributed by Chris Kypridemos.

• New Polya-Gamma Gibbs sampler (method = "pg", now the default): a joint Polya-Gamma data-augmentation sampler with exact full conditionals and no Metropolis tuning. Supports overdispersion, age/period/cohort heterogeneity and period/cohort covariates natively. The legacy Taylor sampler remains available via method = "taylor".
• Native C implementation of the Polya-Gamma sampler (pg_engine = "C", the default) with an equivalent pure-R reference (pg_engine = "R"); the two agree to numerical tolerance.
• mcmc.options values number_of_iterations, burn_in and step may now be set to "auto" (the default), which chooses the MCMC length from the rarity of the data. Any value given as a number is used exactly as before.
• New selectModel(): automatic APC model selection by DIC.
• New prior_scale argument for method = "pg".
• checkConvergence() now assesses the identified quantities (smoothing precisions, intercept and the fitted linear predictor per Lexis cell), which are invariant to the age-period-cohort trend aliasing, rather than the raw effect chains that drift along the non-identified trend.
• effects.apc() and plot.apc() gain a convention argument that fixes the non-identified linear trend to a chosen display gauge, making the effect curves reproducible between runs; plot.apc() also handles any number of quantiles and zero-covariate models.
• print.apc() now also reports the intercept (5%/50%/95% quantiles).
• Fixed predict_apc(periods = 0) crash (downward-sequence off-by-one).
• Fixed predict_apc() logit overflow: use plogis() instead of exp(x)/(1+exp(x)), which returned NaN for large forecast logits and crashed downstream.
• Fixed bamp(): age/period/cohort = NULL is now accepted (previously errored with “argument is of length zero”).
• Fixed predict_apc(): age-period models without a cohort effect can now be predicted, and non-integer population/exposure no longer produces NA.
• bamp(..., method = "pg"): a chain that fails under forked parallelism (parallel::mclapply) now reports its actual error instead of the opaque “subscript out of bounds” that resulted from silently indexing into the failed chain’s result.
• Fixed bamp()’s convergence warning being tied to verbose backwards: it now prints the “did not converge” message when verbose = TRUE and stays silent when verbose = FALSE, as intended.

bamp 2.2.0

• Effects (age, period, cohort) are now computed automatically inside bamp() and stored in the returned object (model$effects), so a separate call to effects.apc() is no longer needed for the default median summary.
• Fixed bug in effects.apc(): cache check was looking for x$effect (singular) instead of x$effects (plural), and referenced an undefined variable in the cache condition.
• MCMC chains now use warm starts: on a restart the previous sample values are used as initial values, reducing burn-in time for automatic convergence checking. First-run behaviour is unchanged.
• Fixed bug in cohort heterogeneity hyperparameter check (cohort="rw2+het" was not recognized correctly).
• Fixed typo in period_covariate handling that silently prevented vector coercion.
• Fixed checkConvergence(): cohort convergence check used age hyperparameter instead of cohort hyperparameter.
• Fixed hyperparameter display in checkConvergence() info output.
• Fixed memory leaks in C++ MCMC code (raw pointer allocations per iteration).
• Fixed GetRNGstate/PutRNGstate pairing in random number generation.
• Added Cholesky decomposition error check with informative message.
• Replaced fork-based mclapply with socket-based makeCluster/parLapply for reliable parallel execution on all platforms, including macOS GUI environments (RStudio). Falls back to sequential if cluster setup fails.
• MCMC chains that fail due to numerical errors are now discarded cleanly instead of continuing with corrupted values.
• Cholesky decomposition now recovers from non-positive-definite precision matrices (common with RW2 priors during tuning) by adding increasing diagonal regularization and retrying, instead of aborting the chain.
• Added unit tests.
• Updated CITATION to use bibentry().

bamp 2.1.5

• bugfix RW2 by Chris Kypridemos
• Fixed population dimension handling in predict_apc.R by pietrosa

bamp 2.1.3

• invalid UTF-8 in comment removed

bamp 2.1.2

• Adapted to R 4.2

bamp 2.1.1

• USE_FC_LEN_T for Fortran code

bamp 2.1.0

• Better default settings (burn in times ten, more informative prior for age).
• Add warnings for failed convergence checks and removed chains in bamp(), including suggestions.
• Add warnings for failed convergence checks in print.apc().
• Fixed unwanted doubling of MCMC when verbose=2.

bamp 2.0.8

• Minor bug fixes, fix “additional issues”.

bamp 2.0.7

• Better initial setting for restarting iterations; helps with RW2 priors.

bamp 2.0.6

• Introductory vignette renamed (double vignette name warning from CRAN).

bamp 2.0.5

• Removed ambiguities (mail Brian Ripley) and clean up in C code.

bamp 2.0.4

• Added examples to all functions.

bamp 2.0.3

• Added more details to help pages.

bamp 2.0.2

• Reference in description changed.

bamp 2.0.1

• Smaller vignettes.

bamp 2.0.0

• R package.