EpiNow2 1.6.0

A release that introduces model improvements to the Gaussian Process models, alongside a number of other improvements and bug fixes.

Documentation

• The documentation of the rt argument has been expanded in the case where rt = NULL to make explicit the settings that are applied in that case. By @jamesmbaazam in #779 and reviewed by @sbfnk.
• The README has been updated to link to the free course on nowcasting and forecasting. The availability of variational inference, Laplace approximation, and Pathfinder through cmdstanr has also be surfaced. By @jamesmbaazam in #753 and reviewed by @seabbs.
• Some implicit argument defaults have been made explicit in the function definition. By @Bisaloo in #729.
• The installation guide in the README has been updated to provide instructions for configuring the C toolchain of Windows, MacOS, and Linux. By @jamesmbaazam in #707 and reviewed by @sbfnk.

Model changes

• The interface for defining delay distributions has been generalised to also cater for continuous distributions
• When defining probability distributions these can now be truncated using the tolerance argument
• Ornstein-Uhlenbeck and 5 / 2 Matérn kernels have been added. By @sbfnk in #741 and reviewed by @seabbs.
• Gaussian processes have been vectorised, leading to some speed gains 🚀 , and the gp_opts() function has gained three more options, "periodic", "ou", and "se", to specify periodic and linear kernels respectively. By @seabbs in #742 and reviewed by @jamesmbaazam.
• Prior predictive checks have been used to update the following priors: the prior on the magnitude of the Gaussian process (from HalfNormal(0, 1) to HalfNormal(0, 0.1)), and the prior on the overdispersion (from 1 / HalfNormal(0, 1)^2 to 1 / HalfNormal(0, 0.25)). In the user-facing API, this is a change in default values of the sd of phi in obs_opts() from 1 to 0.25. By @seabbs in #742 and reviewed by @jamesmbaazam.
• The default stan control options have been updated from list(adapt_delta = 0.95, max_treedepth = 15) to list(adapt_delta = 0.9, max_treedepth = 12) due to improved performance and to reduce the runtime of the default parameterisations. By @seabbs in #742 and reviewed by @jamesmbaazam.
• Initialisation has been simplified by sampling directly from the priors, where possible, rather than from a constrained space. By @seabbs in #742 and reviewed by @jamesmbaazam.
• Unnecessary normalisation of delay priors has been removed. By @seabbs in #742 and reviewed by @jamesmbaazam.
• Ornstein-Uhlenbeck and 5 / 2 Matérn kernels have been added. By @sbfnk in #741 and reviewed by @seabbs.
• Switch to broadcasting from random walks and added unit tests. By @seabbs in #747 and reviewed by @jamesmbaazam.
• Optimised convolution code to take into account the relative length of the vectors being convolved. See #745 by @seabbs and reviewed by @jamesmbaazam.
• Switch to broadcasting the day of the week effect. By @seabbs in #746 and reviewed by @jamesmbaazam.
• A warning is now thrown if nonparametric PMFs passed to delay options have consecutive tail values that are below a certain low threshold as these lead to loss in speed with little gain in accuracy. By @jamesmbaazam in #752 and reviewed by @seabbs, with a subsequent bug fix in #802.
• dist_fit() can now accept any number of samples without throwing a warning when samples < 1000 in #751 by @jamesmbaazam and reviewed by @seabbs and @sbfnk.

Package changes

• epinow() now returns the "timing" output in a "time difference"" format that is easier to understand and work with. By @jamesmbaazam in #688 and reviewed by @sbfnk.
• All functions now use the {cli} R package to signal errors, warnings, and messages. By @jamesmbaazam in #762 and reviewed by @seabbs.
• fix_dist() has been renamed to fix_parameters() because it removes the uncertainty in a distribution's parameters. By @sbfnk in #733 and reviewed by @jamesmbaazam.
• plot.dist_spec now uses color instead of line types to display pmfs vs cmfs. By @jamesmbaazam in #788 and reviewed by @sbfnk.
• The use of the {progressr} package for displaying progress bars is now optional, as is the use of {future} and {future.apply} for parallelisation. By @sbfnk in #798 and reviewed by @seabbs.
• Specifying nonparametric generation time intervals with a nonzero first element (corresponding to the zero bin) is being deprecated as the current behaviour of setting it to zero internally was not well exposed. By @sbfnk in #.

Bug fixes

• a bug was fixed that caused delay option functions to report an error if only the CDF cutoff was specified. By @sbfnk in #716 and reviewed by @jamesmbaazam.
• a bug was fixed where forecast_secondary() did not work with fixed delays. By @sbfnk in #717 and reviewed by @seabbs.
• a bug was fixed that caused delay option functions to report an error if only the CDF cutoff was specified. By @sbfnk.
• a bug was fixed that led to the truncation PMF being shortened from the wrong side when the truncation PMF was longer than the supplied data. By @seabbs in #736 and reviewed by @sbfnk and @jamesmbaazam.
• a bug was fixed that caused internal validation checks on delay distributions to fail if they contained non-parametric distributions. By @jamesmbaazam in #750 and reviewed by @seabbs.
• a bug was fixed where combined distributions where shown in alphabetical order, rather than the order in which they were combined. By @sbfnk in #784 and reviewed by @jamesmbaazam.

Documentation

• Updated the documentation of the dots argument of the stan_sampling_opts() to add that the dots are passed to cmdstanr::sample(). By @jamesmbaazam in #699 and reviewed by @sbfnk.
• generation_time_opts() has been shortened to gt_opts() to make it easier to specify. Calls to both functions are equivalent. By @jamesmbaazam in #698 and reviewed by @seabbs and @sbfnk .
• Added stan documentation for update_rt(). By @seabbs in #747 and reviewed by @jamesmbaazam.