likelihood.model 1.0.1

Bug fix

• fim.likelihood_model no longer forwards the inner ... to the hess_loglik evaluator. DGP-only kwargs (right-censoring time tau, masking probability p, observation functor observe, etc.) flow to rdata only, where they belong. The likelihood layer is computed on the data alone.

  Two motivations for the fix:

  1. Theory: under standard masking condition C3, the masking probability is independent of the system parameters by definition, so the likelihood does not depend on it. The previous behavior leaked a parameter the likelihood is not supposed to see.

  2. R partial-matching: a kwarg named p would silently partial- match the formal par in any hess_loglik method, corrupting the parameter vector at the call site. This made the fim() default fail deterministically for any model whose hess_loglik had par as a formal and any DGP kwarg starting with p.

  Behavioral note: the docstring already documented this contract (“Additional arguments passed to rdata”); the implementation now matches the documentation.

likelihood.model 1.0.0

Breaking Changes

• Removed likelihood_contr_model (R6 contribution-based model builder). This functionality has been extracted to the new likelihood.contr package.
• Removed likelihood_name() (named distribution wrapper). Use likelihood.contr::contr_name() instead.
• Removed weibull_uncensored / likelihood_exact_weibull (Weibull example model). Use likelihood.contr for Weibull likelihood contributions.
• Removed observed_info() generic (was deprecated in v0.10.0). Use -hess_loglik(model)(df, par) or observed_fim() on MLE results.
• Dropped R6 from Imports (no longer needed).
• mvtnorm moved to Suggests (only needed for multivariate sampler()).
• Dropped dplyr and tibble from Suggests.

Notes

• The core framework (loglik, score, hess_loglik, fim, fit, Fisherian inference, lrt, fisher_mle, fisher_boot) is unchanged.
• exponential_lifetime remains as the reference implementation.
• For contribution-based models with heterogeneous observation types, use the likelihood.contr package.

likelihood.model 0.10.0

Breaking Changes

• observed_info() is deprecated; use -hess_loglik(model)(df, par) or observed_fim() on MLE results
• likelihood_exact_weibull() is deprecated in favor of weibull_uncensored()
• Unrecognized censoring values in loglik.likelihood_name_model() now error instead of warning (silently returning 0 produced incorrect MLEs during optimization)
• Removed redundant R6 instance methods $loglik(), $score(), $hess_loglik() from likelihood_contr_model; use the S3 generics loglik(model), score(model), hess_loglik(model) instead
• mvtnorm moved from Imports to Suggests (only needed for multivariate sampler())

Bug Fixes

• Fixed fim.likelihood_model() Monte Carlo estimator: replaced biased outer-product-of-scores with negative expected Hessian approach, which is unbiased at any parameter value
• Fixed fim.likelihood_model() not forwarding ... to hess_loglik, and added NaN/Inf guard to prevent silent corruption of FIM estimates
• Fixed ... not forwarded in likelihood_contr_model numerical fallback for score and hess_loglik
• Fixed parent.frame() in likelihood_contr_model$new() capturing R6 internals instead of the caller’s environment
• Fixed sampler.likelihood_model() ... scoping bug in bootstrap statistic function
• Added optimizer-friendly boundary guards to weibull_uncensored: loglik returns penalty, score/hess_loglik return zeros for invalid parameters (k <= 0 or l <= 0)
• Added input validation in loglik.likelihood_name_model(): errors with a clear message when ob_col is missing from the data frame

New Features

• New rdata.weibull_uncensored() method for generating Weibull data from the model’s DGP
• bias.fisher_mle() now supports optional Monte Carlo bias estimation when theta and model are provided; returns zeros (asymptotic result) otherwise
• mse.fisher_mle() now forwards model and n_sim to bias() for MC-based MSE
• likelihood_contr_model$new() gains envir parameter controlling where loglik.<type>, score.<type>, and hess_loglik.<type> are looked up
• fim.likelihood_model() now adds parameter names to the output matrix when theta is named, and includes tryCatch with informative warnings on partial failures
• evidence() is now an S3 generic, extensible by model classes
• lrt() returns a proper lrt_result class with a print method
• New loglik.likelihood_model and rdata.likelihood_model default methods provide clear error messages when a model class doesn’t implement the required concept methods

likelihood.model 0.9.1

New Features

• Added interval censoring support to likelihood_name() via ob_col_upper parameter
• Input validation in likelihood_name() constructor: checks that distribution functions exist
• Parameter count validation in prepare_args_list() with clear error messages
• Warning on unrecognized censoring values (e.g., typos like “righ”)
• Auto-switch from Nelder-Mead to BFGS for 1-dimensional optimization
• New Fisherian inference functions: support(), relative_likelihood(), likelihood_interval(), profile_loglik(), evidence()
• New exponential_lifetime model with closed-form MLE and right-censoring support
• New fisher_mle and fisher_boot result objects with full S3 method suite

Bug Fixes

• Fixed rowSums error in likelihood_contr_model score for single-parameter models
• Fixed print.likelihood_model crash when ob_col is NULL (e.g., for contr models)

algebraic.mle Integration

• New re-exported generics from algebraic.mle: params(), nparams(), observed_fim(), obs(), mse()
• Methods for fisher_mle objects provide a unified interface across the likelihood ecosystem

Documentation

• New algebraic-mle-integration vignette demonstrating the three-package ecosystem (likelihood.model + algebraic.mle + algebraic.dist)
• Expanded likelihood-name-model vignette: left-censoring, interval censoring, mixed types, Fisherian inference, multiple distributions
• Expanded likelihood-contributions vignette: Case 2 inference, single-parameter example, best practices, interpretive prose for computed results
• New getting-started vignette with comprehensive introduction
• New exponential-lifetime vignette

likelihood.model 0.9.0

• Initial public release with likelihood_name(), likelihood_contr_model, and weibull_uncensored