example_forecaster: Simple example forecaster that forecasts confirmed cases.

In dshemetov/modeltools-mirror: Tools for Building COVID Forecasters and Related Models

Description

Simple example forecaster that forecasts confirmed cases.

Usage

example_forecaster(
  df_list,
  forecast_date,
  ahead = 1:4,
  training_window_size = 14,
  lags = seq(0, -1 * training_window_size, -7),
  quantiles = modeltools::covidhub_probs
)

Arguments

df_list	List of downloaded covidcast signals.
forecast_date	Date from which the forecast should be made.
ahead	Vector of epiweeks ahead to predict.
training_window_size	Number of days prior to forecast_date to use for training.
lags	Vector of lags or list of vectors of lags to apply to df_list. See documentation of dt argument to covidcast::aggregate_signals() for details.
quantiles	Vector of quantile values at which to predict preds <- evalcast::get_predictions(modeltools::example_forecaster, "example", tibble( data_source = "jhu-csse", signal = "confirmed_incidence_num", geo_type = "state", start_date = as.Date("2021-01-01"), ), as.Date("2021-03-01"), "epiweek")

dshemetov/modeltools-mirror documentation built on Jan. 7, 2022, 12:23 a.m.