Skip to contents

Changelog

Source: NEWS.md

ameras (development version)

Bug fixes

  • Fixed an issue with FMA where large differences in AIC values between dose realizations caused NaN weights, resulting in an error.
  • Fixed an issue where a subsequent call to confint() would print intervals for dose-related parameters only. confint() now prints intervals for all parameters.
  • Fixed validation for proportional hazards models specified with Surv(entry, exit, status). ameras() now correctly checks the observed entry and exit time values and errors when any subject has entry > exit.
  • Fixed conditional logistic regression fits with strata() terms that use a matched set column name other than setnr.
  • Fixed BMA handling for models with a single sampled model parameter, including during sample-based confidence interval construction.
  • Fixed an issue where an FMA realization with a finite fitted Hessian but non-finite sampling covariance could trigger a low-level rmvnorm() error. Such realizations are now excluded with a warning before FMA weights and samples are computed.
  • BMA now errors clearly if included.realizations.BMA leaves fewer than two dose realizations, since one-realization analyses should use RC.
  • Conditional logistic regression now explicitly excludes matched sets of size 1 and matched sets with no cases, and errors for matched sets with more than one case.

Improvements

  • FMA no longer uses a cutoff of 20% excluded realizations before warning about potential computational issues. The user is now always warned about the number of excluded realizations.
  • ameras() now supports a na.action argument for model-input missing values after formula terms have been expanded. By default it follows getOption("na.action") (typically na.omit), stores the omitted-row action on the fitted object, and reapplies the same policy when data are supplied later for objects fitted with keep.data=FALSE. na.fail and na.pass are also supported. na.exclude is accepted for fitting, but residuals and diagnostic plots are returned for the fitted rows rather than padded back to the originally supplied row count.
  • FMA no longer uses a minimum sample size of 1 for each realization. Realizations with a model averaging weight yielding a sample size of 0 are excluded. The user is informed of the number of realizations excluded for this reason through a message().
  • Added an additional warning for FMA for the situation when results are based on only 1 or 2-5 realizations after exclusions.
  • FMA realization-specific fits can now use the future framework via future.apply when available. Users can enable parallel execution by setting a future plan before calling ameras(), and future.chunk.size.FMA controls the chunk size passed to future.apply::future_lapply().
  • Right-hand-side covariate terms now support namespace-qualified model-matrix basis functions such as splines::ns() and splines::bs(). For keep.data=FALSE fits, later calls that use supplied data apply the same expanded covariate columns as the original fit.
  • Added numerical diagnostics that warn when right-hand-side covariates appear poorly scaled or ill-conditioned, and when optim() reports convergence but optimizer diagnostics suggest the solution may not be fully stationary. When the Hessian is usable, the optimizer warning uses the approximate remaining objective improvement on both absolute and relative scales.
  • Added convergence() for amerasfit objects to extract or recompute optimizer gradient diagnostics for RC, ERC, and MCML fits.
  • summary() now reports row counts in the order applied during fitting: supplied rows, rows omitted by na.action, and rows used for fitting. For conditional logistic regression, it also reports rows excluded because they belong to uninformative matched sets.
  • Streamlined information printed by summary() and confint(): columns pval.lower and pval.upper for profile likelihood intervals are no longer printed. They are still accessible within the fit object, and warnings are still printed in case an inaccurate profile likelihood bound is suspected.
  • Added structured timing information to fitted method results, separating fitting time, confidence interval computation time, and total time. Printed runtime summaries now use CPU time, so time spent while the computer is asleep is not counted. The existing runtime field is retained as a compatibility summary and is updated when confint() adds confidence interval timing.
  • Added vignettes for standard analyses with one dose realization, manual FMA from realization-specific RC fits, and parallel FMA with the future framework.
  • Added validation to reject input data containing the reserved ameras column name rcdose_ameras.

ameras 0.4.0

CRAN release: 2026-05-29

Bug fixes

  • Fixed a bug causing the inpar check in ameras() to generate an error when it should not.
  • Fixed an issue where FMA generated an error instead of returning NULL for generated samples when all individual fits were excluded.
  • Fixed an issue where setting keep.data=FALSE and passing data to confint() failed a validation check when effect modifiers were present.
  • Fixed an issue where profile likelihood confidence intervals for the ERC method of the proportional hazards family were computed using the non-ERC likelihood, silently ignoring the measurement error correction.
  • Removed the isSymmetric check for FMA variance matrices, which caused platform-dependent differences in included realizations due to numerical differences between Cholesky and solve-based computations. The Cholesky-based variance computation now used guarantees exact symmetry without an explicit check.

Improvements

  • Reduced memory usage and improved computation speed for large datasets: removed the use of N x N matrices for ERC for the proportional hazards family, and precomputed centered dose matrices for ERC are now stored and reused across likelihood evaluations rather than recomputed at each call.
  • confint() no longer recomputes confidence intervals by default if they have already been computed. Use force=TRUE to recompute with different settings. confint() now also prints the computed confidence intervals when called with print=TRUE (default).
  • FMA and BMA output now includes a variance-covariance matrix vcov, computed from the model-averaged posterior samples.
  • It is now possible to use a dosevars variable defined locally (e.g., within a simulation script) through all_of(dosevars) in the formula passed to ameras().

New methods and accessors

  • residuals(): computes Pearson, deviance, and response residuals for all supported families, and Schoenfeld residuals for family="prophaz", supporting both raw and scaled versions following Grambsch and Therneau (1994).
  • plot(): diagnostic plots including residuals versus fitted values and normal Q-Q plots. For family="prophaz", Schoenfeld residual plots are produced to assess the proportional hazards assumption.
  • vcov(): extracts the variance-covariance matrix of parameter estimates for one or more estimation methods.
  • included_realizations(): returns the indices of realizations included in FMA and BMA model averaging.
  • Rhat(): returns the Gelman-Rubin convergence diagnostics and effective sample sizes for BMA results.
  • summary_table(): extracts the summary table from a summary.amerasfit object as a data frame, for programmatic access to parameter estimates, standard errors, and confidence intervals.

ameras 0.3.0

CRAN release: 2026-05-07

Breaking changes

  • Replaced all occurrences of replicate with realization, including in names of arguments and output. As a result, the argument included.replicates.BMA is now deprecated and will be removed in version 1.0.0. Use included.realizations.BMA instead.

New features

  • ecdfplot() now has an argument show.mean (default TRUE) to toggle whether to plot curves for the distribution of the mean doses across realizations and individuals.

Minor improvements and fixes

  • Shortened column names CI.lowerbound and CI.upperbound in the summary() output to CI.lower and CI.upper, respectively.

  • Substantially increased ERC computation speed for the clogit and prophaz families.

ameras 0.2.0

CRAN release: 2026-04-26

Breaking changes

  • Confidence intervals are no longer computed inside ameras(). The arguments CI, params.profCI, maxit.profCI, and tol.profCI are deprecated and will be removed in version 1.0.0. Use the new confint() method instead. See ?confint.amerasfit for details.

  • The direct argument interface to ameras() is deprecated and will be removed in version 1.0.0. The arguments Y, dosevars, doseRRmod, deg, M, X, offset, entry, exit, and setnr are deprecated. Please use the new formula interface instead. See ?ameras for details.

New features

  • Implemented a formula interface for ameras(). The dose variable is specified using the special dose() term, which supports tidyselect syntax for selecting dose columns and allows specifying the dose-response model and effect modifiers directly in the formula. See ?ameras for details and examples.

  • Added confint.amerasfit() for computing confidence intervals separately from model fitting. See ?confint.amerasfit for details.

  • Added function ecdfplot() for exploratory visualization of the dose realizations before model fitting.

General improvements

  • Reduced memory usage for large datasets.

    • Removed the use of an N x N matrix for ERC for the Poisson family, improving both memory and computation speed.
    • Removed duplicate data storage for RC and ERC for all families.

  • summary.amerasfit() now only includes confidence interval columns after they have been computed via confint(). Before calling confint(), a note is printed directing the user to compute confidence intervals.

  • Profile likelihood confidence interval bounds now include p-values in the summary table, making it easier to assess the accuracy of the root-finding algorithm.

  • Removed memoization of the profile likelihood function as it was causing issues and likely not providing much benefit.

New arguments

  • keep.data added to ameras() (default TRUE). When TRUE, the data are stored on the returned amerasfit object, which is required for profile likelihood confidence interval computation via confint() without re-supplying the data. Set to FALSE to reduce memory usage for large datasets, in which case the data must be supplied to confint() explicitly. See ?ameras and ?confint.amerasfit for details.

ameras 0.1.1

CRAN release: 2026-03-29

  • Initial CRAN submission.