Predict cases for a single Rt sample forecasts

Usage

predict_cases(
  cases = NULL,
  rts = NULL,
  serial_interval = NULL,
  forecast_date = NULL,
  horizon = NULL,
  rdist = NULL
)

Arguments

cases: A dataframe containing date and cases variables
rts: A dataframe of containing two variables rt and date with rt being numeric and date being a date.
serial_interval: A numeric vector describing the probability distribution the serial interval. See EpiNow::covid_serial_interval for an example of the format.
forecast_date: A character string date (format "yyyy-mm-dd") indicating the forecast date. Defaults to NULL in which case it will be assumed that the forecast date is the last data present in the cases dataframe
horizon: Numeric, the time horizon over which to predict.
rdist: A function to be used to sample the number of cases. Must take two arguments with the first specfying the number of samples and the second the mean. Defaults to rpois if not supplied

Value

Forecast cases for over a future forecast horizon

Examples


forecast <- forecast_rt(EpiSoon::example_obs_rts[1:10, ],
  model = function(...) {
    EpiSoon::bsts_model(
      model =
        function(ss, y) {
          bsts::AddSemilocalLinearTrend(ss, y = y)
        }, ...
    )
  },
  horizon = 7, samples = 1
)


purrr::map_dfr(1:100, ~ predict_cases(
  cases = EpiSoon::example_obs_cases,
  rts = forecast,
  forecast_date = as.Date("2020-03-10"),
  serial_interval = example_serial_interval
)) %>%
  dplyr::group_by(date) %>%
  dplyr::summarise(cases = mean(cases))
#> # A tibble: 7 × 2
#>   date       cases
#>   <date>     <dbl>
#> 1 2020-03-11  168.
#> 2 2020-03-12  201.
#> 3 2020-03-13  233.
#> 4 2020-03-14  272.
#> 5 2020-03-15  314.
#> 6 2020-03-16  363.
#> 7 2020-03-17  408.