#> Warning: ! No training data was supplied to `prep()`.
#> ! Unlike a <recipe>, an <epi_recipe> does not
#> ! store the full template data in the object.
#> ! Please supply the training data to the `prep()` function,
#> ! to avoid addtional warning messages.
#> An `epi_df` object, 20,496 x 8 with metadata:
#> * geo_type = state
#> * time_type = day
#> * as_of = 2022-05-31 19:08:25.791826
#>
#> # A tibble: 20,496 × 8
#> geo_value time_value case_rate death_rate lag_diff_7_case_rate
#> * <chr> <date> <dbl> <dbl> <dbl>
#> 1 ak 2020-12-31 35.9 0.158 NA
#> 2 al 2020-12-31 65.1 0.438 NA
#> 3 ar 2020-12-31 66.0 1.27 NA
#> 4 as 2020-12-31 0 0 NA
#> 5 az 2020-12-31 76.8 1.10 NA
#> 6 ca 2020-12-31 96.0 0.751 NA
#> 7 co 2020-12-31 35.8 0.649 NA
#> 8 ct 2020-12-31 52.1 0.819 NA
#> 9 dc 2020-12-31 31.0 0.601 NA
#> 10 de 2020-12-31 65.2 0.807 NA
#> # ℹ 20,486 more rows
#> # ℹ 3 more variables: lag_diff_14_case_rate <dbl>, lag_diff_7_death_rate <dbl>,
#> # lag_diff_14_death_rate <dbl>
This example should filter out
NA's so it demonstrates the behavior better https://cmu-delphi.github.io/epipredict/reference/step_lag_difference.html