NEWS.md

aedseo (development version)

Features

• The aedseo package now allows for modelling of binomial data. to_time_series() now accepts samples and proportion as arguments. The proportion argument is the proportion of cases in the total tested samples at every time point. These variables can be used in the seasonal_onset() function with the family arguments binomial or quasibinomial and in the seasonal_burden_levels() function with the family argument beta. Attributes; outcome_type, model_outcome and burden_outcome have been added to the object classes of to_time_series(), seasonal_onset() and seasonal_burden_levels() functions to always be able to track if the results are based on count/incidence or binomial data (#100, #101, #103).
• summary.tsd_onset() now reports proportional observations on the proportion scale and prints an explicit NA reference when no seasonal onset was detected, instead of producing no output (#103).

Fixes

• The estimate_disease_threshold() function defaulted to 1 if the selected observation was below 1. Since the package has the possibility to use incidence as an input there is a need for allowing values below 1. This was fixed in #106.

aedseo 1.1.0

Features

• combined_seasonal_output() now includes a seasonal_offset logical variables in the output that estimates when the season has ended after the first seasonal_onset. It can be controlled by the inputs: burden_level_decrease and steps_with_decrease (#93).

aedseo 1.0.1

Fixes

• fit_growth_rate() would incorrectly determine confidence intervals when using ATLAS BLAS/LAPACK (#89).

aedseo 1.0.0

Deprecations

Features

• Added new arguments incidence and incidence_denominator to to_time_series() that allows the user to get output as incidence (#84).

• Added new argument population to to_time_series() and fit_growth_rate() that allows the user to add the background population connected to each observation (#83).

• Added new argument use_offset to seasonal_onset() that allows the user to add the background population to adjust the growth rate estimations (#83).

• Added new feature to estimate multiple waves in combined_seasonal_output (#77).

• Added estimate_disease_threshold() for users to easier estimate the disease specific threshold (#85).

Improvements

• Observations are now divided into cases and incidence, which is implemented into all functions in the package. Cases are used as default, but if the user additionally inputs population the output will be incidence (#84).

Minor changes

aedseo 0.3.0

Deprecations

• aedseo() is now deprecated. Please use seasonal_onset() instead. A warning is shown when using aedseo() (#41).

• tsd() is now deprecated. Please use to_time_series() instead. A warning is shown when using tsd() (#41).

Features

• Added the seasonal_burden_levels() function, which calculates burden levels based on data from previous seasons with two different methods; "peak_levels" or "intensity_levels" (#37).

• Added the fit_percentiles() function, which optimises a user selected distribution and calculates the percentiles based on observations and weights. It is meant to be used within the seasonal_burden_levels() function (#35, #37) - Renamed fit_quantiles() to fit_percentiles() (#60).

• Added combined_seasonal_output() as the main function to run both seasonal_onset() and seasonal_burden_levels() to get a combined result for the newest season (#44).

• Added consecutive_growth_warnings() function to help the user with a method to define the disease-specific threshold (#80).

• Added a new argument only_current_season to seasonal_onset(), seasonal_burden_levels() and combined_seasonal_output() which gives the possibility to either get output from only the current season or for all available seasons (#45).

• Added historical_summary() which uses a tsd_onset object to summarise historical estimations (#75).

• summary() can now summarise tsd_burden_level objects (#60).

• plot() and autoplot() can now plot tsd_combined_seasonal_output and tsd_consecutive_growth_warning objects (#57, #80).

• Added generate_seasonal_data() to generate synthetic data for testing and documentation purposes (#56).

• Added seasonal_onset() as a replacement for the deprecated aedseo() function (#41).

• Added to_time_series() as a replacement for the deprecated tsd() function (#41).

Improvements

• Enhanced clarity and user guidance in the vignettes:

  • vignette("generate_seasonal_wave"),
  • vignette("aedseo"),
  • vignette("seasonal_onset")
  • vignette("burden_levels") providing a comprehensive walkthrough of the application of the functions provided by the aedseo package with detailed explanations and illustrative examples (#56, #57, #58, #59, #60, #61).

• Improved the autoplot() function which can now visualise dates as days, weeks and months on the x-axis with the time_interval argument (#56).

• Improved the epi_calendar() function to work for a season spanning new year (#34).

• Using predict() on tsd_onset objects now uses the same time-scale as the given object (#61). That is, the time_interval attribute controls if predictions are by "days", "weeks" or "months".

• The aedseo() function now allows for the choice of adding season as an input argument (#34).

• {checkmate} assertions have been added to enhance user feedback with clearer error messages and to ensure functions operate correctly by validating inputs (#33).

• Improved the aedseo() function to work with NA values. The user now defines how many NA values the function should allow in each window (#32).

Minor changes

• Added Sofia Myrup Otero as an author of the R package (#55).

• Added Rasmus Skytte Randløv as a reviewer of the R package (#55).

• The disease_threshold argument now reflects the disease threshold in one time step. If the total number of cases in a window of size k exceeds disease_threshold * k, a seasonal onset alarm can be triggered (#32).

aedseo 0.1.2

Minor changes

• Transferring maintainership of the R package to Lasse Engbo Christiansen.

aedseo 0.1.1

Improvements

• Enhanced clarity and user guidance in the introductory vignette, providing a more comprehensive walkthrough of the application of the 'aeddo' algorithm on time series data with detailed explanations and illustrative examples.

Minor changes

• Updated LICENSE.md to have Statens Serum Institut as a copyright holder.

• Fixed installation guide for the development version in the README.Rmd and README.md

• Added Lasse Engbo Christiansen as an author of the R package.

• Added a new function epi_calendar() that determines the epidemiological season based on a given date, allowing users to easily categorize dates within or outside specified seasons.

• Introduced additional visualizations in the autoplot() method, enhancing the capabilities of the plot() method with new displays of observed cases and growth rates.

aedseo 0.1.0

Features

• Added the aedseo function, which automates the early detection of seasonal epidemic onsets by estimating growth rates for consecutive time intervals and calculating the Sum of Cases (sum_of_cases).

• Introduced autoplot and plot methods for visualizing aedseo and aedseo_tsd objects. These functions allow you to create insightful ggplot2 plots for your data.

• Included the fit_growth_rate function, enabling users to fit growth rate models to time series observations.

• Introduced the predict method for aedseo objects, which allows you to predict observations for future time steps given the growth rates.

• Added the summary method for aedseo objects, providing a comprehensive summary of the results.

• Introduced the tsd function, allowing users to create S3 aedseo_tsd (time-series data) objects from observed data and corresponding dates.