- Simplifies and optimises the internal functions used to estimate the parametric daily reporting probability. These are now exposed to the user via the
distributionparameter with both the Lognormal and Gamma families being tested to work. Note that both parameterisations use their standard parameterisations as given in the stan manual (see #42 by @adrian-lison and @seabbs)
- Add profiling switch to model compilation, allowing to toggle profiling (https://mc-stan.org/cmdstanr/articles/profiling.html) on/off in the same model. Also supports .stan files found in
include_paths(see #41 and #54 by @adrian-lison).
- Fully vectorise the likelihood by flattening observations and pre-specify expected observations into a vector before calculating the log-likelihood (see #40 by @seabbs).
- Adds vectorisation of zero truncated normal distributions (see #38 by @seabbs)
hazard_to_probhas been optimised using vectorisation (see [#53] by @adrian-lison and @seabbs).
prob_to_hazardhas been optimised so that only required cumulative probabilties are calculated (see [#53] by @adrian-lison and @seabbs).
- Convert retrospective data date fields to class of
enw_retrospective_datato solve esoteric error.
- Added full argument name for
include_pathsto avoid console chatter
- Adds a
enw_model()and specifies a new default of
list("01")which enables simple pre-compilation optimisations. See here of these optimisatiosn for details.
logitas may instead use base R
- Add support for extracting and summarising posterior nowcast samples
- Package spell check
- Update read me quick start to use 40 days of delay vs 30
- Add a section to the read me quick start showing an example of handling nowcast samples.
- Add support for passing custom models and included files to
- Fix a bug where
enw_summarise_samples()returned duplicate samples.
- Add support for passing holidays as a variable and then adjusting by converting the holiday day into a custom day of the week (by default Sunday but this is set by the user).
- Added support for scoring on both the natural and log scale. This represents absolute and relative scoring respectively.
- Add support for passing in priors
- Add case study vignette
- Add model definition and implementation details.
- Add support for out of sample scoring (using
- Initial version of the package with broadly working functionality and first draft vignettes.