diff --git a/episodes/data/covid_150days.rds b/episodes/data/covid_150days.rds
new file mode 100644
index 00000000..3c46c240
Binary files /dev/null and b/episodes/data/covid_150days.rds differ
diff --git a/episodes/data/covid_300days.rds b/episodes/data/covid_300days.rds
new file mode 100644
index 00000000..ff00e0e2
Binary files /dev/null and b/episodes/data/covid_300days.rds differ
diff --git a/episodes/data/covid_490days.rds b/episodes/data/covid_490days.rds
new file mode 100644
index 00000000..3bf32db1
Binary files /dev/null and b/episodes/data/covid_490days.rds differ
diff --git a/instructors/data-covid-filter.R b/instructors/data-covid-filter.R
new file mode 100644
index 00000000..bb44b122
--- /dev/null
+++ b/instructors/data-covid-filter.R
@@ -0,0 +1,49 @@
+
+library(tidyverse)
+library(incidence2)
+
+# read data ---------------------------------------------------------------
+
+incidence_class <- incidence2::covidregionaldataUK %>%
+  as_tibble() %>%
+  filter(region == "London") %>%
+  # preprocess missing values
+  tidyr::replace_na(
+    list(
+      deaths_new = 0,
+      cases_new = 0
+    )
+  ) %>%
+  # compute the daily incidence
+  incidence2::incidence(
+    date_index = "date",
+    counts = c("cases_new","deaths_new"),
+    date_names_to = "date",
+    complete_dates = TRUE
+  ) %>%
+  identity()
+
+
+incidence_class %>%
+  # filter(date < ymd(20200415)) %>%
+  # filter(date < ymd(20200701)) %>%
+  plot()
+
+covid_incidence2 <- incidence_class %>%
+  cfr::prepare_data(
+    cases_variable = "cases_new",
+    deaths_variable = "deaths_new") %>%
+  as_tibble()
+
+covid_incidence2
+
+covid_incidence2 %>%
+  write_rds(file.path("episodes", "data", "covid_490days.rds"))
+
+covid_incidence2 %>%
+  filter(date < ymd(20200415)) %>%
+  write_rds(file.path("episodes", "data", "covid_150days.rds"))
+
+covid_incidence2 %>%
+  filter(date < ymd(20200701)) %>%
+  write_rds(file.path("episodes", "data", "covid_300days.rds"))