STATS 220

STATS 220Don’t repeat yourself 🐈1 / 38

2 / 38

Case study: batch download and process COVID-19 data

3 / 38

Keep calm, and import CSVs one by one

library(tidyverse) # library(purrr)
covid_0301 <- read_csv("https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/03-01-2020.csv")
covid_0302 <- read_csv("https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/03-02-2020.csv")
covid_0303 <- read_csv("https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/03-03-2020.csv")
(covid_03 <- bind_rows(covid_0301, covid_0302, covid_0303))

#> # A tibble: 420 x 8
#>   `Province/State` `Country/Region` `Last Update`      
#>   <chr>            <chr>            <dttm>             
#> 1 Hubei            Mainland China   2020-03-01 10:13:19
#> 2 <NA>             South Korea      2020-03-01 23:43:03
#> 3 <NA>             Italy            2020-03-01 23:23:02
#> 4 Guangdong        Mainland China   2020-03-01 14:13:18
#> 5 Henan            Mainland China   2020-03-01 14:13:18
#> 6 Zhejiang         Mainland China   2020-03-01 10:13:33
#> # … with 414 more rows, and 5 more variables:
#> #   Confirmed <dbl>, Deaths <dbl>, Recovered <dbl>,
#> #   Latitude <dbl>, Longitude <dbl>

4 / 38

Data is changing

covid_1231 <- read_csv("https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/12-31-2020.csv")
covid_1231

#> # A tibble: 3,980 x 14
#>    FIPS Admin2 Province_State Country_Region     
#>   <dbl> <chr>  <chr>          <chr>              
#> 1    NA <NA>   <NA>           Afghanistan        
#> 2    NA <NA>   <NA>           Albania            
#> 3    NA <NA>   <NA>           Algeria            
#> 4    NA <NA>   <NA>           Andorra            
#> 5    NA <NA>   <NA>           Angola             
#> 6    NA <NA>   <NA>           Antigua and Barbuda
#> # … with 3,974 more rows, and 10 more variables:
#> #   Last_Update <dttm>, Lat <dbl>, Long_ <dbl>,
#> #   Confirmed <dbl>, Deaths <dbl>, Recovered <dbl>,
#> #   Active <dbl>, Combined_Key <chr>, Incident_Rate <dbl>,
#> #   Case_Fatality_Ratio <dbl>

5 / 38

What are the repeated patterns?everyday different URLs
read each csv file into R
non-standardised column names
DRY: automate the common tasks above6 / 38

1 generate a sequence of urls

make_url <- function(date) {
  date_chr <- format(date, "%m-%d-%Y")
  setNames(glue::glue("https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/{date_chr}.csv"),
  date_chr)
}

3 key steps to creating a custom function:

pick a name for the function, e.g. make_url.
list the inputs, or arguments, to the function inside function, e.g. date.
place the code you have developed in body of the function, a { block that immediately follows function(...).

7 / 38

Test `make_url()`: does it return what we expected?

library(lubridate)
dates <- seq(mdy("03-01-2020"), mdy("12-31-2020"), by = 1)
urls <- make_url(dates)
head(urls, 2)

#> https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/03-01-2020.csv
#> https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/03-02-2020.csv

tail(urls, 2)

#> https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/12-30-2020.csv
#> https://github.com/CSSEGISandData/COVID-19/raw/master/csse_covid_19_data/csse_covid_19_daily_reports/12-31-2020.csv

8 / 38

2 & 3 read data and standardise column names

read_covid19 <- function(url) {
  data <- read_csv(url, 
    col_types = cols(Last_Update = "_", `Last Update` = "_"))
  rename_with(data, janitor::make_clean_names)
}
read_covid19(urls[1])

#> # A tibble: 126 x 7
#>   province_state country_region confirmed deaths recovered
#>   <chr>          <chr>              <dbl>  <dbl>     <dbl>
#> 1 Hubei          Mainland China     66907   2761     31536
#> 2 <NA>           South Korea         3736     17        30
#> 3 <NA>           Italy               1694     34        83
#> 4 Guangdong      Mainland China      1349      7      1016
#> 5 Henan          Mainland China      1272     22      1198
#> 6 Zhejiang       Mainland China      1205      1      1046
#> # … with 120 more rows, and 2 more variables:
#> #   latitude <dbl>, longitude <dbl>

9 / 38

Why functionModularise the code to make it easier to understand the logic.
Generalise the code to make it reusable later.
Reduce the duplication and make it less error-prone.
10 / 38

The joy of functional programming (FP)

Hadley Wickham: The joy of functional programming

11 / 38

comes to rescue

copy and paste

covid19_csv1 <- read_covid19(urls[1])
covid19_csv2 <- read_covid19(urls[2])
covid19_csv3 <- read_covid19(urls[2])
covid19_csv4 <- read_covid19(urls[4])
covid19_csv5 <- read_covid19(urls[5])
covid19_csv6 <- read_covid19(urls[6])
covid19_csv6 <- read_covid19(urls[7])
covid19_csv8 <- read_covid19(urls[8])
covid19_csv9 <- read_covid19(urls[9])

When it's gonna end? 😭

The goal of FP is to make it easy to express repeated actions using high-level verbs.

12 / 38

introduce typos/errors, when just simply copy and paste

- `map()`

map(.x, .f, ...) for every element of .x, apply .f

always returns a list

covid19_lst <- map(urls, read_covid19)
covid19_lst[[length(covid19_lst)]]

#> # A tibble: 3,980 x 13
#>    fips admin2 province_state country_region      lat   long
#>   <dbl> <chr>  <chr>          <chr>             <dbl>  <dbl>
#> 1    NA <NA>   <NA>           Afghanistan        33.9  67.7 
#> 2    NA <NA>   <NA>           Albania            41.2  20.2 
#> 3    NA <NA>   <NA>           Algeria            28.0   1.66
#> 4    NA <NA>   <NA>           Andorra            42.5   1.52
#> 5    NA <NA>   <NA>           Angola            -11.2  17.9 
#> 6    NA <NA>   <NA>           Antigua and Barb…  17.1 -61.8 
#> # … with 3,974 more rows, and 7 more variables:
#> #   confirmed <dbl>, deaths <dbl>, recovered <dbl>,
#> #   active <dbl>, combined_key <chr>, incident_rate <dbl>,
#> #   case_fatality_ratio <dbl>

13 / 38

- `map()`

A functional is a function that takes a function as an input and returns a vector as output.

14 / 38

- `map()`

.f takes a function, formula, or vector.

If a function, it is used as is.
If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:
- For a single argument function, use .
- For a two argument function, use .x and .y
- For more arguments, use ..1, ..2, ..3 etc
If a character vector index by name and a numeric vector index by position.

covid19_lst <- map(urls, read_covid19)
covid19_lst <- map(urls, ~ read_covid19(.x))
covid19_lst <- map(urls, function(.x) read_covid19(.x))

15 / 38

- `map()`

image credit: Jenny Bryan

15 / 38

- `map()`

image credit: Jenny Bryan

15 / 38

- `map()`

image credit: Jenny Bryan

15 / 38

- `map()`

Bind multiple tibbles by row

bind_rows(covid19_lst)

#> # A tibble: 1,069,141 x 16
#>   province_state country_region confirmed deaths recovered
#>   <chr>          <chr>              <dbl>  <dbl>     <dbl>
#> 1 Hubei          Mainland China     66907   2761     31536
#> 2 <NA>           South Korea         3736     17        30
#> 3 <NA>           Italy               1694     34        83
#> 4 Guangdong      Mainland China      1349      7      1016
#> 5 Henan          Mainland China      1272     22      1198
#> 6 Zhejiang       Mainland China      1205      1      1046
#> # … with 1,069,135 more rows, and 11 more variables:
#> #   latitude <dbl>, longitude <dbl>, fips <dbl>,
#> #   admin2 <chr>, lat <dbl>, long <dbl>, active <dbl>,
#> #   combined_key <chr>, incidence_rate <dbl>,
#> #   case_fatality_ratio <dbl>, incident_rate <dbl>

16 / 38

- `map()`

Hmm, where are the dates?

bind_rows(covid19_lst)

#> # A tibble: 1,069,141 x 16
#>   province_state country_region confirmed deaths recovered
#>   <chr>          <chr>              <dbl>  <dbl>     <dbl>
#> 1 Hubei          Mainland China     66907   2761     31536
#> 2 <NA>           South Korea         3736     17        30
#> 3 <NA>           Italy               1694     34        83
#> 4 Guangdong      Mainland China      1349      7      1016
#> 5 Henan          Mainland China      1272     22      1198
#> 6 Zhejiang       Mainland China      1205      1      1046
#> # … with 1,069,135 more rows, and 11 more variables:
#> #   latitude <dbl>, longitude <dbl>, fips <dbl>,
#> #   admin2 <chr>, lat <dbl>, long <dbl>, active <dbl>,
#> #   combined_key <chr>, incidence_rate <dbl>,
#> #   case_fatality_ratio <dbl>, incident_rate <dbl>

17 / 38

- `map()`

- `map2()`

map2(.x, .y, .f) for every element of .x and .y, apply .f

covid19_lst <- map2(urls, dates,
  ~ read_covid19(.x) %>% mutate(date = .y))
bind_rows(covid19_lst)

#> # A tibble: 1,069,141 x 17
#>   province_state country_region confirmed deaths recovered
#>   <chr>          <chr>              <dbl>  <dbl>     <dbl>
#> 1 Hubei          Mainland China     66907   2761     31536
#> 2 <NA>           South Korea         3736     17        30
#> 3 <NA>           Italy               1694     34        83
#> 4 Guangdong      Mainland China      1349      7      1016
#> 5 Henan          Mainland China      1272     22      1198
#> 6 Zhejiang       Mainland China      1205      1      1046
#> # … with 1,069,135 more rows, and 12 more variables:
#> #   latitude <dbl>, longitude <dbl>, fips <dbl>,
#> #   admin2 <chr>, lat <dbl>, long <dbl>, active <dbl>,
#> #   combined_key <chr>, incidence_rate <dbl>,
#> #   case_fatality_ratio <dbl>, incident_rate <dbl>,
#> #   date <date>

18 / 38

- `map()`

- `map2()`

19 / 38

- `map()`

- `map2()`

- `map_dfr()`

map_*() is a variant of map() that returns an output in the indicated type. map_dfr() returns data frames by row-binding.

covid19_raw <- map_dfr(urls, read_covid19, .id = "date")
covid19_raw

#> # A tibble: 1,069,141 x 17
#>   date       province_state country_region confirmed deaths
#>   <chr>      <chr>          <chr>              <dbl>  <dbl>
#> 1 03-01-2020 Hubei          Mainland China     66907   2761
#> 2 03-01-2020 <NA>           South Korea         3736     17
#> 3 03-01-2020 <NA>           Italy               1694     34
#> 4 03-01-2020 Guangdong      Mainland China      1349      7
#> 5 03-01-2020 Henan          Mainland China      1272     22
#> 6 03-01-2020 Zhejiang       Mainland China      1205      1
#> # … with 1,069,135 more rows, and 12 more variables:
#> #   recovered <dbl>, latitude <dbl>, longitude <dbl>,
#> #   fips <dbl>, admin2 <chr>, lat <dbl>, long <dbl>,
#> #   active <dbl>, combined_key <chr>, incidence_rate <dbl>,
#> #   case_fatality_ratio <dbl>, incident_rate <dbl>

20 / 38

- `map()`

- `map2()`

- `map_dfr()`

- `map_*()`

map() variants

map(.x, .f) maps over 1 argument returns a list
map_dbl(.x, .f) returns a double vector
map_chr(.x, .f) returns a character vector
map_lgl(.x, .f) returns a logical vector
map_dfr(.x, .f) returns a data frame created by row-binding
map_dfc(.x, .f) returns a data frame created by col-binding
map2_*(.x, .y, .f) maps over 2 arguments
pmap_*(.l, .f) maps over > 2 arguments

21 / 38

22 / 38

Wrangle a bit

covid19_bystate <- covid19_raw %>% 
  select(date, country_region:recovered) %>% 
  mutate(
    date = mdy(date),
    country_region = case_when(
      country_region == "Korea, South" ~ "South Korea",
      country_region == "Mainland China" ~ "China",
      TRUE ~ country_region)
  )
covid19_bystate

#> # A tibble: 1,069,141 x 5
#>   date       country_region confirmed deaths recovered
#>   <date>     <chr>              <dbl>  <dbl>     <dbl>
#> 1 2020-03-01 China              66907   2761     31536
#> 2 2020-03-01 South Korea         3736     17        30
#> 3 2020-03-01 Italy               1694     34        83
#> 4 2020-03-01 China               1349      7      1016
#> 5 2020-03-01 China               1272     22      1198
#> 6 2020-03-01 China               1205      1      1046
#> # … with 1,069,135 more rows

23 / 38

covid19_byregion <- covid19_bystate %>% 
  group_by(country_region, date) %>% 
  summarise(
    confirmed = sum(confirmed, na.rm = TRUE),
    deaths = sum(deaths, na.rm = TRUE),
    recovered = sum(recovered, na.rm = TRUE)
  ) %>% 
  mutate(
    new_confirmed = confirmed - lag(confirmed, na.rm = TRUE),
    new_deaths = deaths - lag(deaths, na.rm = TRUE),
    new_recovered = recovered - lag(recovered, na.rm = TRUE)
  ) %>% 
  ungroup()

Don't wanna type 😭, but I typed...

24 / 38

Apply a function across() multiple columnscovid19_byregion <- covid19_bystate %>% 
  group_by(country_region, date) %>% 
  summarise(
    across(
      confirmed:recovered,
      sum, na.rm = TRUE)
  ) %>% 
  mutate(
    across(
      confirmed:recovered,
      ~ .x - lag(.x, na.rm = TRUE),
      .names = "new_{.col}")
  ) %>% 
  ungroup()

covid19_byregion <- covid19_bystate %>% 
  group_by(country_region, date) %>% 
  summarise(
    across(
      where(is.numeric),
      sum, na.rm = TRUE)
  ) %>% 
  mutate(
    across(
      where(is.numeric),
      ~ .x - lag(.x, na.rm = TRUE),
      .names = "new_{.col}")
  ) %>% 
  ungroup()

25 / 38

covid19_byregion

#> # A tibble: 56,107 x 8
#>   country_region date       confirmed deaths recovered
#>   <chr>          <date>         <dbl>  <dbl>     <dbl>
#> 1 Afghanistan    2020-03-01         1      0         0
#> 2 Afghanistan    2020-03-02         1      0         0
#> 3 Afghanistan    2020-03-03         2      0         0
#> 4 Afghanistan    2020-03-04         4      0         0
#> 5 Afghanistan    2020-03-05         4      0         0
#> 6 Afghanistan    2020-03-06         4      0         0
#> # … with 56,101 more rows, and 3 more variables:
#> #   new_confirmed <dbl>, new_deaths <dbl>,
#> #   new_recovered <dbl>

26 / 38

nest() multiple columns into a list-column

covid19_lstcol <- covid19_byregion %>% 
  nest(data = -country_region)
covid19_lstcol

#> # A tibble: 236 x 2
#>   country_region      data                  
#>   <chr>               <list>                
#> 1 Afghanistan         <tibble[,7] [306 × 7]>
#> 2 Albania             <tibble[,7] [298 × 7]>
#> 3 Algeria             <tibble[,7] [306 × 7]>
#> 4 Andorra             <tibble[,7] [305 × 7]>
#> 5 Angola              <tibble[,7] [287 × 7]>
#> 6 Antigua and Barbuda <tibble[,7] [294 × 7]>
#> # … with 230 more rows

27 / 38

Look inside the list-column

covid19_au <- covid19_lstcol$data[[10]]
covid19_au

#> # A tibble: 306 x 7
#>   date       confirmed deaths recovered new_confirmed
#>   <date>         <dbl>  <dbl>     <dbl>         <dbl>
#> 1 2020-03-01        27      1        11            NA
#> 2 2020-03-02        30      1        11             3
#> 3 2020-03-03        39      1        11             9
#> 4 2020-03-04        52      2        11            13
#> 5 2020-03-05        55      2        21             3
#> 6 2020-03-06        60      2        21             5
#> # … with 300 more rows, and 2 more variables:
#> #   new_deaths <dbl>, new_recovered <dbl>

28 / 38

Fit regression models and extract slopes

fit_au <- lm(new_recovered ~ lag(new_confirmed, n = 14), 
  data = covid19_au)
fit_au

#> 
#> Call:
#> lm(formula = new_recovered ~ lag(new_confirmed, n = 14), data = covid19_au)
#> 
#> Coefficients:
#>                (Intercept)  lag(new_confirmed, n = 14)  
#>                    26.3548                      0.5299

slope_au <- coef(fit_au)[2]
slope_au

#> lag(new_confirmed, n = 14) 
#>                  0.5299203

29 / 38

Function time

extract_slope <- function(data) {
  fit <- lm(new_recovered ~ lag(new_confirmed, n = 14), data)
  coef(fit)[2]
}

Oops, map() isn't happy. 🙁

covid19_lstcol$data %>% 
  map(extract_slope)

#> Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases

30 / 38

If an error occured, safely() captures the error

returns a named list of result and error

safely_extract_slope <- safely(extract_slope, 
  otherwise = NA_real_)
covid19_lstcol$data %>% map(safely_extract_slope)

#> [[1]]
#> [[1]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6920523 
#> 
#> [[1]]$error
#> NULL
#> 
#> 
#> [[2]]
#> [[2]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6435756 
#> 
#> [[2]]$error
#> NULL
#> 
#> 
#> [[3]]
#> [[3]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5329582 
#> 
#> [[3]]$error
#> NULL
#> 
#> 
#> [[4]]
#> [[4]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5595175 
#> 
#> [[4]]$error
#> NULL
#> 
#> 
#> [[5]]
#> [[5]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5196389 
#> 
#> [[5]]$error
#> NULL
#> 
#> 
#> [[6]]
#> [[6]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7375993 
#> 
#> [[6]]$error
#> NULL
#> 
#> 
#> [[7]]
#> [[7]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8669078 
#> 
#> [[7]]$error
#> NULL
#> 
#> 
#> [[8]]
#> [[8]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8971555 
#> 
#> [[8]]$error
#> NULL
#> 
#> 
#> [[9]]
#> [[9]]$result
#> [1] NA
#> 
#> [[9]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[10]]
#> [[10]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5299203 
#> 
#> [[10]]$error
#> NULL
#> 
#> 
#> [[11]]
#> [[11]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9182422 
#> 
#> [[11]]$error
#> NULL
#> 
#> 
#> [[12]]
#> [[12]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9642226 
#> 
#> [[12]]$error
#> NULL
#> 
#> 
#> [[13]]
#> [[13]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2903094 
#> 
#> [[13]]$error
#> NULL
#> 
#> 
#> [[14]]
#> [[14]]$result
#> [1] NA
#> 
#> [[14]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[15]]
#> [[15]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8760548 
#> 
#> [[15]]$error
#> NULL
#> 
#> 
#> [[16]]
#> [[16]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8306039 
#> 
#> [[16]]$error
#> NULL
#> 
#> 
#> [[17]]
#> [[17]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1478383 
#> 
#> [[17]]$error
#> NULL
#> 
#> 
#> [[18]]
#> [[18]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.038319 
#> 
#> [[18]]$error
#> NULL
#> 
#> 
#> [[19]]
#> [[19]]$result
#> lag(new_confirmed, n = 14) 
#>                 -0.1417372 
#> 
#> [[19]]$error
#> NULL
#> 
#> 
#> [[20]]
#> [[20]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.3760244 
#> 
#> [[20]]$error
#> NULL
#> 
#> 
#> [[21]]
#> [[21]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2665434 
#> 
#> [[21]]$error
#> NULL
#> 
#> 
#> [[22]]
#> [[22]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.075549 
#> 
#> [[22]]$error
#> NULL
#> 
#> 
#> [[23]]
#> [[23]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6108023 
#> 
#> [[23]]$error
#> NULL
#> 
#> 
#> [[24]]
#> [[24]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6446241 
#> 
#> [[24]]$error
#> NULL
#> 
#> 
#> [[25]]
#> [[25]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.491272 
#> 
#> [[25]]$error
#> NULL
#> 
#> 
#> [[26]]
#> [[26]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6540097 
#> 
#> [[26]]$error
#> NULL
#> 
#> 
#> [[27]]
#> [[27]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4138478 
#> 
#> [[27]]$error
#> NULL
#> 
#> 
#> [[28]]
#> [[28]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.573295 
#> 
#> [[28]]$error
#> NULL
#> 
#> 
#> [[29]]
#> [[29]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6704973 
#> 
#> [[29]]$error
#> NULL
#> 
#> 
#> [[30]]
#> [[30]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8998348 
#> 
#> [[30]]$error
#> NULL
#> 
#> 
#> [[31]]
#> [[31]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.05736442 
#> 
#> [[31]]$error
#> NULL
#> 
#> 
#> [[32]]
#> [[32]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5809416 
#> 
#> [[32]]$error
#> NULL
#> 
#> 
#> [[33]]
#> [[33]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1256359 
#> 
#> [[33]]$error
#> NULL
#> 
#> 
#> [[34]]
#> [[34]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1167159 
#> 
#> [[34]]$error
#> NULL
#> 
#> 
#> [[35]]
#> [[35]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9418862 
#> 
#> [[35]]$error
#> NULL
#> 
#> 
#> [[36]]
#> [[36]]$result
#> [1] NA
#> 
#> [[36]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[37]]
#> [[37]]$result
#> [1] NA
#> 
#> [[37]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[38]]
#> [[38]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2580678 
#> 
#> [[38]]$error
#> NULL
#> 
#> 
#> [[39]]
#> [[39]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4058524 
#> 
#> [[39]]$error
#> NULL
#> 
#> 
#> [[40]]
#> [[40]]$result
#> [1] NA
#> 
#> [[40]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[41]]
#> [[41]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4110214 
#> 
#> [[41]]$error
#> NULL
#> 
#> 
#> [[42]]
#> [[42]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8410514 
#> 
#> [[42]]$error
#> NULL
#> 
#> 
#> [[43]]
#> [[43]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6855367 
#> 
#> [[43]]$error
#> NULL
#> 
#> 
#> [[44]]
#> [[44]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1374103 
#> 
#> [[44]]$error
#> NULL
#> 
#> 
#> [[45]]
#> [[45]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2689623 
#> 
#> [[45]]$error
#> NULL
#> 
#> 
#> [[46]]
#> [[46]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5826228 
#> 
#> [[46]]$error
#> NULL
#> 
#> 
#> [[47]]
#> [[47]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.004124 
#> 
#> [[47]]$error
#> NULL
#> 
#> 
#> [[48]]
#> [[48]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6459742 
#> 
#> [[48]]$error
#> NULL
#> 
#> 
#> [[49]]
#> [[49]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8822818 
#> 
#> [[49]]$error
#> NULL
#> 
#> 
#> [[50]]
#> [[50]]$result
#> [1] NA
#> 
#> [[50]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[51]]
#> [[51]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7600437 
#> 
#> [[51]]$error
#> NULL
#> 
#> 
#> [[52]]
#> [[52]]$result
#> [1] NA
#> 
#> [[52]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[53]]
#> [[53]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.01675946 
#> 
#> [[53]]$error
#> NULL
#> 
#> 
#> [[54]]
#> [[54]]$result
#> [1] NA
#> 
#> [[54]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[55]]
#> [[55]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9635528 
#> 
#> [[55]]$error
#> NULL
#> 
#> 
#> [[56]]
#> [[56]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.049798 
#> 
#> [[56]]$error
#> NULL
#> 
#> 
#> [[57]]
#> [[57]]$result
#> lag(new_confirmed, n = 14) 
#>                         NA 
#> 
#> [[57]]$error
#> NULL
#> 
#> 
#> [[58]]
#> [[58]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5009491 
#> 
#> [[58]]$error
#> NULL
#> 
#> 
#> [[59]]
#> [[59]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1780097 
#> 
#> [[59]]$error
#> NULL
#> 
#> 
#> [[60]]
#> [[60]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6290575 
#> 
#> [[60]]$error
#> NULL
#> 
#> 
#> [[61]]
#> [[61]]$result
#> [1] NA
#> 
#> [[61]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[62]]
#> [[62]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.05348175 
#> 
#> [[62]]$error
#> NULL
#> 
#> 
#> [[63]]
#> [[63]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2492007 
#> 
#> [[63]]$error
#> NULL
#> 
#> 
#> [[64]]
#> [[64]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5076202 
#> 
#> [[64]]$error
#> NULL
#> 
#> 
#> [[65]]
#> [[65]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.01289268 
#> 
#> [[65]]$error
#> NULL
#> 
#> 
#> [[66]]
#> [[66]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1628999 
#> 
#> [[66]]$error
#> NULL
#> 
#> 
#> [[67]]
#> [[67]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8211096 
#> 
#> [[67]]$error
#> NULL
#> 
#> 
#> [[68]]
#> [[68]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6890107 
#> 
#> [[68]]$error
#> NULL
#> 
#> 
#> [[69]]
#> [[69]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4747831 
#> 
#> [[69]]$error
#> NULL
#> 
#> 
#> [[70]]
#> [[70]]$result
#> [1] NA
#> 
#> [[70]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[71]]
#> [[71]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1370787 
#> 
#> [[71]]$error
#> NULL
#> 
#> 
#> [[72]]
#> [[72]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8498793 
#> 
#> [[72]]$error
#> NULL
#> 
#> 
#> [[73]]
#> [[73]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.02613858 
#> 
#> [[73]]$error
#> NULL
#> 
#> 
#> [[74]]
#> [[74]]$result
#> [1] NA
#> 
#> [[74]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[75]]
#> [[75]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4039257 
#> 
#> [[75]]$error
#> NULL
#> 
#> 
#> [[76]]
#> [[76]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2783185 
#> 
#> [[76]]$error
#> NULL
#> 
#> 
#> [[77]]
#> [[77]]$result
#> [1] NA
#> 
#> [[77]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[78]]
#> [[78]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8992036 
#> 
#> [[78]]$error
#> NULL
#> 
#> 
#> [[79]]
#> [[79]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.748486 
#> 
#> [[79]]$error
#> NULL
#> 
#> 
#> [[80]]
#> [[80]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4874482 
#> 
#> [[80]]$error
#> NULL
#> 
#> 
#> [[81]]
#> [[81]]$result
#> [1] NA
#> 
#> [[81]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[82]]
#> [[82]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9351811 
#> 
#> [[82]]$error
#> NULL
#> 
#> 
#> [[83]]
#> [[83]]$result
#> [1] NA
#> 
#> [[83]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[84]]
#> [[84]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.02809673 
#> 
#> [[84]]$error
#> NULL
#> 
#> 
#> [[85]]
#> [[85]]$result
#> [1] NA
#> 
#> [[85]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[86]]
#> [[86]]$result
#> [1] NA
#> 
#> [[86]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[87]]
#> [[87]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8144507 
#> 
#> [[87]]$error
#> NULL
#> 
#> 
#> [[88]]
#> [[88]]$result
#> [1] NA
#> 
#> [[88]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[89]]
#> [[89]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2512326 
#> 
#> [[89]]$error
#> NULL
#> 
#> 
#> [[90]]
#> [[90]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.01763243 
#> 
#> [[90]]$error
#> NULL
#> 
#> 
#> [[91]]
#> [[91]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5299206 
#> 
#> [[91]]$error
#> NULL
#> 
#> 
#> [[92]]
#> [[92]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4224714 
#> 
#> [[92]]$error
#> NULL
#> 
#> 
#> [[93]]
#> [[93]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.01196319 
#> 
#> [[93]]$error
#> NULL
#> 
#> 
#> [[94]]
#> [[94]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2954188 
#> 
#> [[94]]$error
#> NULL
#> 
#> 
#> [[95]]
#> [[95]]$result
#> [1] NA
#> 
#> [[95]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[96]]
#> [[96]]$result
#> [1] NA
#> 
#> [[96]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[97]]
#> [[97]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4821869 
#> 
#> [[97]]$error
#> NULL
#> 
#> 
#> [[98]]
#> [[98]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7901768 
#> 
#> [[98]]$error
#> NULL
#> 
#> 
#> [[99]]
#> [[99]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8979504 
#> 
#> [[99]]$error
#> NULL
#> 
#> 
#> [[100]]
#> [[100]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9487975 
#> 
#> [[100]]$error
#> NULL
#> 
#> 
#> [[101]]
#> [[101]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7877892 
#> 
#> [[101]]$error
#> NULL
#> 
#> 
#> [[102]]
#> [[102]]$result
#> [1] NA
#> 
#> [[102]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[103]]
#> [[103]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8682548 
#> 
#> [[103]]$error
#> NULL
#> 
#> 
#> [[104]]
#> [[104]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2087615 
#> 
#> [[104]]$error
#> NULL
#> 
#> 
#> [[105]]
#> [[105]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8144906 
#> 
#> [[105]]$error
#> NULL
#> 
#> 
#> [[106]]
#> [[106]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7282967 
#> 
#> [[106]]$error
#> NULL
#> 
#> 
#> [[107]]
#> [[107]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5431636 
#> 
#> [[107]]$error
#> NULL
#> 
#> 
#> [[108]]
#> [[108]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8678327 
#> 
#> [[108]]$error
#> NULL
#> 
#> 
#> [[109]]
#> [[109]]$result
#> [1] NA
#> 
#> [[109]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[110]]
#> [[110]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.177228 
#> 
#> [[110]]$error
#> NULL
#> 
#> 
#> [[111]]
#> [[111]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1624997 
#> 
#> [[111]]$error
#> NULL
#> 
#> 
#> [[112]]
#> [[112]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7411408 
#> 
#> [[112]]$error
#> NULL
#> 
#> 
#> [[113]]
#> [[113]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.02179744 
#> 
#> [[113]]$error
#> NULL
#> 
#> 
#> [[114]]
#> [[114]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9123329 
#> 
#> [[114]]$error
#> NULL
#> 
#> 
#> [[115]]
#> [[115]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1588501 
#> 
#> [[115]]$error
#> NULL
#> 
#> 
#> [[116]]
#> [[116]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.0223889 
#> 
#> [[116]]$error
#> NULL
#> 
#> 
#> [[117]]
#> [[117]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9392607 
#> 
#> [[117]]$error
#> NULL
#> 
#> 
#> [[118]]
#> [[118]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7137065 
#> 
#> [[118]]$error
#> NULL
#> 
#> 
#> [[119]]
#> [[119]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1059572 
#> 
#> [[119]]$error
#> NULL
#> 
#> 
#> [[120]]
#> [[120]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1280151 
#> 
#> [[120]]$error
#> NULL
#> 
#> 
#> [[121]]
#> [[121]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7095778 
#> 
#> [[121]]$error
#> NULL
#> 
#> 
#> [[122]]
#> [[122]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8902059 
#> 
#> [[122]]$error
#> NULL
#> 
#> 
#> [[123]]
#> [[123]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8601565 
#> 
#> [[123]]$error
#> NULL
#> 
#> 
#> [[124]]
#> [[124]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.933333 
#> 
#> [[124]]$error
#> NULL
#> 
#> 
#> [[125]]
#> [[125]]$result
#> [1] NA
#> 
#> [[125]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[126]]
#> [[126]]$result
#> [1] NA
#> 
#> [[126]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[127]]
#> [[127]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7684239 
#> 
#> [[127]]$error
#> NULL
#> 
#> 
#> [[128]]
#> [[128]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2407153 
#> 
#> [[128]]$error
#> NULL
#> 
#> 
#> [[129]]
#> [[129]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8741184 
#> 
#> [[129]]$error
#> NULL
#> 
#> 
#> [[130]]
#> [[130]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8370697 
#> 
#> [[130]]$error
#> NULL
#> 
#> 
#> [[131]]
#> [[131]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4458898 
#> 
#> [[131]]$error
#> NULL
#> 
#> 
#> [[132]]
#> [[132]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8722886 
#> 
#> [[132]]$error
#> NULL
#> 
#> 
#> [[133]]
#> [[133]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.01612903 
#> 
#> [[133]]$error
#> NULL
#> 
#> 
#> [[134]]
#> [[134]]$result
#> [1] NA
#> 
#> [[134]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[135]]
#> [[135]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7781136 
#> 
#> [[135]]$error
#> NULL
#> 
#> 
#> [[136]]
#> [[136]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4789353 
#> 
#> [[136]]$error
#> NULL
#> 
#> 
#> [[137]]
#> [[137]]$result
#> [1] NA
#> 
#> [[137]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[138]]
#> [[138]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5295625 
#> 
#> [[138]]$error
#> NULL
#> 
#> 
#> [[139]]
#> [[139]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8784718 
#> 
#> [[139]]$error
#> NULL
#> 
#> 
#> [[140]]
#> [[140]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4523111 
#> 
#> [[140]]$error
#> NULL
#> 
#> 
#> [[141]]
#> [[141]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.06500141 
#> 
#> [[141]]$error
#> NULL
#> 
#> 
#> [[142]]
#> [[142]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7416119 
#> 
#> [[142]]$error
#> NULL
#> 
#> 
#> [[143]]
#> [[143]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8947562 
#> 
#> [[143]]$error
#> NULL
#> 
#> 
#> [[144]]
#> [[144]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7780699 
#> 
#> [[144]]$error
#> NULL
#> 
#> 
#> [[145]]
#> [[145]]$result
#> lag(new_confirmed, n = 14) 
#>               -0.003802281 
#> 
#> [[145]]$error
#> NULL
#> 
#> 
#> [[146]]
#> [[146]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.044826 
#> 
#> [[146]]$error
#> NULL
#> 
#> 
#> [[147]]
#> [[147]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9334443 
#> 
#> [[147]]$error
#> NULL
#> 
#> 
#> [[148]]
#> [[148]]$result
#> lag(new_confirmed, n = 14) 
#>                0.005462774 
#> 
#> [[148]]$error
#> NULL
#> 
#> 
#> [[149]]
#> [[149]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7127593 
#> 
#> [[149]]$error
#> NULL
#> 
#> 
#> [[150]]
#> [[150]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6934141 
#> 
#> [[150]]$error
#> NULL
#> 
#> 
#> [[151]]
#> [[151]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4480632 
#> 
#> [[151]]$error
#> NULL
#> 
#> 
#> [[152]]
#> [[152]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7526479 
#> 
#> [[152]]$error
#> NULL
#> 
#> 
#> [[153]]
#> [[153]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6662816 
#> 
#> [[153]]$error
#> NULL
#> 
#> 
#> [[154]]
#> [[154]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2028236 
#> 
#> [[154]]$error
#> NULL
#> 
#> 
#> [[155]]
#> [[155]]$result
#> [1] NA
#> 
#> [[155]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[156]]
#> [[156]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9219172 
#> 
#> [[156]]$error
#> NULL
#> 
#> 
#> [[157]]
#> [[157]]$result
#> [1] NA
#> 
#> [[157]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[158]]
#> [[158]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5813632 
#> 
#> [[158]]$error
#> NULL
#> 
#> 
#> [[159]]
#> [[159]]$result
#> [1] NA
#> 
#> [[159]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[160]]
#> [[160]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7849124 
#> 
#> [[160]]$error
#> NULL
#> 
#> 
#> [[161]]
#> [[161]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.0817926 
#> 
#> [[161]]$error
#> NULL
#> 
#> 
#> [[162]]
#> [[162]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7567328 
#> 
#> [[162]]$error
#> NULL
#> 
#> 
#> [[163]]
#> [[163]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4486768 
#> 
#> [[163]]$error
#> NULL
#> 
#> 
#> [[164]]
#> [[164]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9711064 
#> 
#> [[164]]$error
#> NULL
#> 
#> 
#> [[165]]
#> [[165]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8404479 
#> 
#> [[165]]$error
#> NULL
#> 
#> 
#> [[166]]
#> [[166]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8340365 
#> 
#> [[166]]$error
#> NULL
#> 
#> 
#> [[167]]
#> [[167]]$result
#> [1] NA
#> 
#> [[167]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[168]]
#> [[168]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.146428 
#> 
#> [[168]]$error
#> NULL
#> 
#> 
#> [[169]]
#> [[169]]$result
#> [1] NA
#> 
#> [[169]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[170]]
#> [[170]]$result
#> [1] NA
#> 
#> [[170]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[171]]
#> [[171]]$result
#> [1] NA
#> 
#> [[171]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[172]]
#> [[172]]$result
#> [1] NA
#> 
#> [[172]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[173]]
#> [[173]]$result
#> [1] NA
#> 
#> [[173]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[174]]
#> [[174]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9711078 
#> 
#> [[174]]$error
#> NULL
#> 
#> 
#> [[175]]
#> [[175]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9645341 
#> 
#> [[175]]$error
#> NULL
#> 
#> 
#> [[176]]
#> [[176]]$result
#> [1] NA
#> 
#> [[176]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[177]]
#> [[177]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1992464 
#> 
#> [[177]]$error
#> NULL
#> 
#> 
#> [[178]]
#> [[178]]$result
#> [1] NA
#> 
#> [[178]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[179]]
#> [[179]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.05725861 
#> 
#> [[179]]$error
#> NULL
#> 
#> 
#> [[180]]
#> [[180]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.5481366 
#> 
#> [[180]]$error
#> NULL
#> 
#> 
#> [[181]]
#> [[181]]$result
#> [1] NA
#> 
#> [[181]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[182]]
#> [[182]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.05410779 
#> 
#> [[182]]$error
#> NULL
#> 
#> 
#> [[183]]
#> [[183]]$result
#> lag(new_confirmed, n = 14) 
#>               -0.007462687 
#> 
#> [[183]]$error
#> NULL
#> 
#> 
#> [[184]]
#> [[184]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.3680185 
#> 
#> [[184]]$error
#> NULL
#> 
#> 
#> [[185]]
#> [[185]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.00138019 
#> 
#> [[185]]$error
#> NULL
#> 
#> 
#> [[186]]
#> [[186]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8943568 
#> 
#> [[186]]$error
#> NULL
#> 
#> 
#> [[187]]
#> [[187]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4699798 
#> 
#> [[187]]$error
#> NULL
#> 
#> 
#> [[188]]
#> [[188]]$result
#> lag(new_confirmed, n = 14) 
#>               -0.001714412 
#> 
#> [[188]]$error
#> NULL
#> 
#> 
#> [[189]]
#> [[189]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.01177239 
#> 
#> [[189]]$error
#> NULL
#> 
#> 
#> [[190]]
#> [[190]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4459296 
#> 
#> [[190]]$error
#> NULL
#> 
#> 
#> [[191]]
#> [[191]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.741075 
#> 
#> [[191]]$error
#> NULL
#> 
#> 
#> [[192]]
#> [[192]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7394529 
#> 
#> [[192]]$error
#> NULL
#> 
#> 
#> [[193]]
#> [[193]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9754129 
#> 
#> [[193]]$error
#> NULL
#> 
#> 
#> [[194]]
#> [[194]]$result
#> lag(new_confirmed, n = 14) 
#>               -0.004582951 
#> 
#> [[194]]$error
#> NULL
#> 
#> 
#> [[195]]
#> [[195]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1134682 
#> 
#> [[195]]$error
#> NULL
#> 
#> 
#> [[196]]
#> [[196]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8579614 
#> 
#> [[196]]$error
#> NULL
#> 
#> 
#> [[197]]
#> [[197]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7022731 
#> 
#> [[197]]$error
#> NULL
#> 
#> 
#> [[198]]
#> [[198]]$result
#> lag(new_confirmed, n = 14) 
#>               0.0003349323 
#> 
#> [[198]]$error
#> NULL
#> 
#> 
#> [[199]]
#> [[199]]$result
#> lag(new_confirmed, n = 14) 
#>                -0.01591102 
#> 
#> [[199]]$error
#> NULL
#> 
#> 
#> [[200]]
#> [[200]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.902218 
#> 
#> [[200]]$error
#> NULL
#> 
#> 
#> [[201]]
#> [[201]]$result
#> [1] NA
#> 
#> [[201]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[202]]
#> [[202]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2367338 
#> 
#> [[202]]$error
#> NULL
#> 
#> 
#> [[203]]
#> [[203]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6937724 
#> 
#> [[203]]$error
#> NULL
#> 
#> 
#> [[204]]
#> [[204]]$result
#> lag(new_confirmed, n = 14) 
#>               8.226172e-05 
#> 
#> [[204]]$error
#> NULL
#> 
#> 
#> [[205]]
#> [[205]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.7049068 
#> 
#> [[205]]$error
#> NULL
#> 
#> 
#> [[206]]
#> [[206]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4773192 
#> 
#> [[206]]$error
#> NULL
#> 
#> 
#> [[207]]
#> [[207]]$result
#> [1] NA
#> 
#> [[207]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[208]]
#> [[208]]$result
#> [1] NA
#> 
#> [[208]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[209]]
#> [[209]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2683079 
#> 
#> [[209]]$error
#> NULL
#> 
#> 
#> [[210]]
#> [[210]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.3773631 
#> 
#> [[210]]$error
#> NULL
#> 
#> 
#> [[211]]
#> [[211]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.09471127 
#> 
#> [[211]]$error
#> NULL
#> 
#> 
#> [[212]]
#> [[212]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.791494 
#> 
#> [[212]]$error
#> NULL
#> 
#> 
#> [[213]]
#> [[213]]$result
#> [1] NA
#> 
#> [[213]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[214]]
#> [[214]]$result
#> [1] NA
#> 
#> [[214]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[215]]
#> [[215]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9127807 
#> 
#> [[215]]$error
#> NULL
#> 
#> 
#> [[216]]
#> [[216]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.6683978 
#> 
#> [[216]]$error
#> NULL
#> 
#> 
#> [[217]]
#> [[217]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4777528 
#> 
#> [[217]]$error
#> NULL
#> 
#> 
#> [[218]]
#> [[218]]$result
#> lag(new_confirmed, n = 14) 
#>                   0.289256 
#> 
#> [[218]]$error
#> NULL
#> 
#> 
#> [[219]]
#> [[219]]$result
#> lag(new_confirmed, n = 14) 
#>                 0.09288966 
#> 
#> [[219]]$error
#> NULL
#> 
#> 
#> [[220]]
#> [[220]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.1472352 
#> 
#> [[220]]$error
#> NULL
#> 
#> 
#> [[221]]
#> [[221]]$result
#> [1] NA
#> 
#> [[221]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[222]]
#> [[222]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8942314 
#> 
#> [[222]]$error
#> NULL
#> 
#> 
#> [[223]]
#> [[223]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.8124493 
#> 
#> [[223]]$error
#> NULL
#> 
#> 
#> [[224]]
#> [[224]]$result
#> lag(new_confirmed, n = 14) 
#>                0.001282686 
#> 
#> [[224]]$error
#> NULL
#> 
#> 
#> [[225]]
#> [[225]]$result
#> lag(new_confirmed, n = 14) 
#>                   1.164049 
#> 
#> [[225]]$error
#> NULL
#> 
#> 
#> [[226]]
#> [[226]]$result
#> lag(new_confirmed, n = 14) 
#>                 -0.7243288 
#> 
#> [[226]]$error
#> NULL
#> 
#> 
#> [[227]]
#> [[227]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9300546 
#> 
#> [[227]]$error
#> NULL
#> 
#> 
#> [[228]]
#> [[228]]$result
#> lag(new_confirmed, n = 14) 
#>                         NA 
#> 
#> [[228]]$error
#> NULL
#> 
#> 
#> [[229]]
#> [[229]]$result
#> [1] NA
#> 
#> [[229]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[230]]
#> [[230]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9314963 
#> 
#> [[230]]$error
#> NULL
#> 
#> 
#> [[231]]
#> [[231]]$result
#> [1] NA
#> 
#> [[231]]$error
#> <simpleError in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): 0 (non-NA) cases>
#> 
#> 
#> [[232]]
#> [[232]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.3004533 
#> 
#> [[232]]$error
#> NULL
#> 
#> 
#> [[233]]
#> [[233]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.9587314 
#> 
#> [[233]]$error
#> NULL
#> 
#> 
#> [[234]]
#> [[234]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.2780126 
#> 
#> [[234]]$error
#> NULL
#> 
#> 
#> [[235]]
#> [[235]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4839777 
#> 
#> [[235]]$error
#> NULL
#> 
#> 
#> [[236]]
#> [[236]]$result
#> lag(new_confirmed, n = 14) 
#>                  0.4202595 
#> 
#> [[236]]$error
#> NULL

31 / 38

Pull out resultscovid19_lstcol$data %>% 
  map(safely_extract_slope) %>% 
  map("result")

#> [[1]]
#> lag(new_confirmed, n = 14) 
#>                  0.6920523 
#> 
#> [[2]]
#> lag(new_confirmed, n = 14) 
#>                  0.6435756 
#> 
#> [[3]]
#> lag(new_confirmed, n = 14) 
#>                  0.5329582 
#> 
#> [[4]]
#> lag(new_confirmed, n = 14) 
#>                  0.5595175 
#> 
#> [[5]]
#> lag(new_confirmed, n = 14) 
#>                  0.5196389 
#> 
#> [[6]]
#> lag(new_confirmed, n = 14) 
#>                  0.7375993 
#> 
#> [[7]]
#> lag(new_confirmed, n = 14) 
#>                  0.8669078 
#> 
#> [[8]]
#> lag(new_confirmed, n = 14) 
#>                  0.8971555 
#> 
#> [[9]]
#> [1] NA
#> 
#> [[10]]
#> lag(new_confirmed, n = 14) 
#>                  0.5299203 
#> 
#> [[11]]
#> lag(new_confirmed, n = 14) 
#>                  0.9182422 
#> 
#> [[12]]
#> lag(new_confirmed, n = 14) 
#>                  0.9642226 
#> 
#> [[13]]
#> lag(new_confirmed, n = 14) 
#>                  0.2903094 
#> 
#> [[14]]
#> [1] NA
#> 
#> [[15]]
#> lag(new_confirmed, n = 14) 
#>                  0.8760548 
#> 
#> [[16]]
#> lag(new_confirmed, n = 14) 
#>                  0.8306039 
#> 
#> [[17]]
#> lag(new_confirmed, n = 14) 
#>                  0.1478383 
#> 
#> [[18]]
#> lag(new_confirmed, n = 14) 
#>                   1.038319 
#> 
#> [[19]]
#> lag(new_confirmed, n = 14) 
#>                 -0.1417372 
#> 
#> [[20]]
#> lag(new_confirmed, n = 14) 
#>                  0.3760244 
#> 
#> [[21]]
#> lag(new_confirmed, n = 14) 
#>                  0.2665434 
#> 
#> [[22]]
#> lag(new_confirmed, n = 14) 
#>                   0.075549 
#> 
#> [[23]]
#> lag(new_confirmed, n = 14) 
#>                  0.6108023 
#> 
#> [[24]]
#> lag(new_confirmed, n = 14) 
#>                  0.6446241 
#> 
#> [[25]]
#> lag(new_confirmed, n = 14) 
#>                   0.491272 
#> 
#> [[26]]
#> lag(new_confirmed, n = 14) 
#>                  0.6540097 
#> 
#> [[27]]
#> lag(new_confirmed, n = 14) 
#>                  0.4138478 
#> 
#> [[28]]
#> lag(new_confirmed, n = 14) 
#>                   0.573295 
#> 
#> [[29]]
#> lag(new_confirmed, n = 14) 
#>                  0.6704973 
#> 
#> [[30]]
#> lag(new_confirmed, n = 14) 
#>                  0.8998348 
#> 
#> [[31]]
#> lag(new_confirmed, n = 14) 
#>                 0.05736442 
#> 
#> [[32]]
#> lag(new_confirmed, n = 14) 
#>                  0.5809416 
#> 
#> [[33]]
#> lag(new_confirmed, n = 14) 
#>                  0.1256359 
#> 
#> [[34]]
#> lag(new_confirmed, n = 14) 
#>                  0.1167159 
#> 
#> [[35]]
#> lag(new_confirmed, n = 14) 
#>                  0.9418862 
#> 
#> [[36]]
#> [1] NA
#> 
#> [[37]]
#> [1] NA
#> 
#> [[38]]
#> lag(new_confirmed, n = 14) 
#>                  0.2580678 
#> 
#> [[39]]
#> lag(new_confirmed, n = 14) 
#>                  0.4058524 
#> 
#> [[40]]
#> [1] NA
#> 
#> [[41]]
#> lag(new_confirmed, n = 14) 
#>                  0.4110214 
#> 
#> [[42]]
#> lag(new_confirmed, n = 14) 
#>                  0.8410514 
#> 
#> [[43]]
#> lag(new_confirmed, n = 14) 
#>                  0.6855367 
#> 
#> [[44]]
#> lag(new_confirmed, n = 14) 
#>                  0.1374103 
#> 
#> [[45]]
#> lag(new_confirmed, n = 14) 
#>                  0.2689623 
#> 
#> [[46]]
#> lag(new_confirmed, n = 14) 
#>                  0.5826228 
#> 
#> [[47]]
#> lag(new_confirmed, n = 14) 
#>                   1.004124 
#> 
#> [[48]]
#> lag(new_confirmed, n = 14) 
#>                  0.6459742 
#> 
#> [[49]]
#> lag(new_confirmed, n = 14) 
#>                  0.8822818 
#> 
#> [[50]]
#> [1] NA
#> 
#> [[51]]
#> lag(new_confirmed, n = 14) 
#>                  0.7600437 
#> 
#> [[52]]
#> [1] NA
#> 
#> [[53]]
#> lag(new_confirmed, n = 14) 
#>                -0.01675946 
#> 
#> [[54]]
#> [1] NA
#> 
#> [[55]]
#> lag(new_confirmed, n = 14) 
#>                  0.9635528 
#> 
#> [[56]]
#> lag(new_confirmed, n = 14) 
#>                   1.049798 
#> 
#> [[57]]
#> lag(new_confirmed, n = 14) 
#>                         NA 
#> 
#> [[58]]
#> lag(new_confirmed, n = 14) 
#>                  0.5009491 
#> 
#> [[59]]
#> lag(new_confirmed, n = 14) 
#>                  0.1780097 
#> 
#> [[60]]
#> lag(new_confirmed, n = 14) 
#>                  0.6290575 
#> 
#> [[61]]
#> [1] NA
#> 
#> [[62]]
#> lag(new_confirmed, n = 14) 
#>                 0.05348175 
#> 
#> [[63]]
#> lag(new_confirmed, n = 14) 
#>                  0.2492007 
#> 
#> [[64]]
#> lag(new_confirmed, n = 14) 
#>                  0.5076202 
#> 
#> [[65]]
#> lag(new_confirmed, n = 14) 
#>                -0.01289268 
#> 
#> [[66]]
#> lag(new_confirmed, n = 14) 
#>                  0.1628999 
#> 
#> [[67]]
#> lag(new_confirmed, n = 14) 
#>                  0.8211096 
#> 
#> [[68]]
#> lag(new_confirmed, n = 14) 
#>                  0.6890107 
#> 
#> [[69]]
#> lag(new_confirmed, n = 14) 
#>                  0.4747831 
#> 
#> [[70]]
#> [1] NA
#> 
#> [[71]]
#> lag(new_confirmed, n = 14) 
#>                  0.1370787 
#> 
#> [[72]]
#> lag(new_confirmed, n = 14) 
#>                  0.8498793 
#> 
#> [[73]]
#> lag(new_confirmed, n = 14) 
#>                 0.02613858 
#> 
#> [[74]]
#> [1] NA
#> 
#> [[75]]
#> lag(new_confirmed, n = 14) 
#>                  0.4039257 
#> 
#> [[76]]
#> lag(new_confirmed, n = 14) 
#>                  0.2783185 
#> 
#> [[77]]
#> [1] NA
#> 
#> [[78]]
#> lag(new_confirmed, n = 14) 
#>                  0.8992036 
#> 
#> [[79]]
#> lag(new_confirmed, n = 14) 
#>                   0.748486 
#> 
#> [[80]]
#> lag(new_confirmed, n = 14) 
#>                  0.4874482 
#> 
#> [[81]]
#> [1] NA
#> 
#> [[82]]
#> lag(new_confirmed, n = 14) 
#>                  0.9351811 
#> 
#> [[83]]
#> [1] NA
#> 
#> [[84]]
#> lag(new_confirmed, n = 14) 
#>                 0.02809673 
#> 
#> [[85]]
#> [1] NA
#> 
#> [[86]]
#> [1] NA
#> 
#> [[87]]
#> lag(new_confirmed, n = 14) 
#>                  0.8144507 
#> 
#> [[88]]
#> [1] NA
#> 
#> [[89]]
#> lag(new_confirmed, n = 14) 
#>                  0.2512326 
#> 
#> [[90]]
#> lag(new_confirmed, n = 14) 
#>                 0.01763243 
#> 
#> [[91]]
#> lag(new_confirmed, n = 14) 
#>                  0.5299206 
#> 
#> [[92]]
#> lag(new_confirmed, n = 14) 
#>                  0.4224714 
#> 
#> [[93]]
#> lag(new_confirmed, n = 14) 
#>                -0.01196319 
#> 
#> [[94]]
#> lag(new_confirmed, n = 14) 
#>                  0.2954188 
#> 
#> [[95]]
#> [1] NA
#> 
#> [[96]]
#> [1] NA
#> 
#> [[97]]
#> lag(new_confirmed, n = 14) 
#>                  0.4821869 
#> 
#> [[98]]
#> lag(new_confirmed, n = 14) 
#>                  0.7901768 
#> 
#> [[99]]
#> lag(new_confirmed, n = 14) 
#>                  0.8979504 
#> 
#> [[100]]
#> lag(new_confirmed, n = 14) 
#>                  0.9487975 
#> 
#> [[101]]
#> lag(new_confirmed, n = 14) 
#>                  0.7877892 
#> 
#> [[102]]
#> [1] NA
#> 
#> [[103]]
#> lag(new_confirmed, n = 14) 
#>                  0.8682548 
#> 
#> [[104]]
#> lag(new_confirmed, n = 14) 
#>                  0.2087615 
#> 
#> [[105]]
#> lag(new_confirmed, n = 14) 
#>                  0.8144906 
#> 
#> [[106]]
#> lag(new_confirmed, n = 14) 
#>                  0.7282967 
#> 
#> [[107]]
#> lag(new_confirmed, n = 14) 
#>                  0.5431636 
#> 
#> [[108]]
#> lag(new_confirmed, n = 14) 
#>                  0.8678327 
#> 
#> [[109]]
#> [1] NA
#> 
#> [[110]]
#> lag(new_confirmed, n = 14) 
#>                   1.177228 
#> 
#> [[111]]
#> lag(new_confirmed, n = 14) 
#>                  0.1624997 
#> 
#> [[112]]
#> lag(new_confirmed, n = 14) 
#>                  0.7411408 
#> 
#> [[113]]
#> lag(new_confirmed, n = 14) 
#>                 0.02179744 
#> 
#> [[114]]
#> lag(new_confirmed, n = 14) 
#>                  0.9123329 
#> 
#> [[115]]
#> lag(new_confirmed, n = 14) 
#>                  0.1588501 
#> 
#> [[116]]
#> lag(new_confirmed, n = 14) 
#>                  0.0223889 
#> 
#> [[117]]
#> lag(new_confirmed, n = 14) 
#>                  0.9392607 
#> 
#> [[118]]
#> lag(new_confirmed, n = 14) 
#>                  0.7137065 
#> 
#> [[119]]
#> lag(new_confirmed, n = 14) 
#>                  0.1059572 
#> 
#> [[120]]
#> lag(new_confirmed, n = 14) 
#>                  0.1280151 
#> 
#> [[121]]
#> lag(new_confirmed, n = 14) 
#>                  0.7095778 
#> 
#> [[122]]
#> lag(new_confirmed, n = 14) 
#>                  0.8902059 
#> 
#> [[123]]
#> lag(new_confirmed, n = 14) 
#>                  0.8601565 
#> 
#> [[124]]
#> lag(new_confirmed, n = 14) 
#>                   0.933333 
#> 
#> [[125]]
#> [1] NA
#> 
#> [[126]]
#> [1] NA
#> 
#> [[127]]
#> lag(new_confirmed, n = 14) 
#>                  0.7684239 
#> 
#> [[128]]
#> lag(new_confirmed, n = 14) 
#>                  0.2407153 
#> 
#> [[129]]
#> lag(new_confirmed, n = 14) 
#>                  0.8741184 
#> 
#> [[130]]
#> lag(new_confirmed, n = 14) 
#>                  0.8370697 
#> 
#> [[131]]
#> lag(new_confirmed, n = 14) 
#>                  0.4458898 
#> 
#> [[132]]
#> lag(new_confirmed, n = 14) 
#>                  0.8722886 
#> 
#> [[133]]
#> lag(new_confirmed, n = 14) 
#>                -0.01612903 
#> 
#> [[134]]
#> [1] NA
#> 
#> [[135]]
#> lag(new_confirmed, n = 14) 
#>                  0.7781136 
#> 
#> [[136]]
#> lag(new_confirmed, n = 14) 
#>                  0.4789353 
#> 
#> [[137]]
#> [1] NA
#> 
#> [[138]]
#> lag(new_confirmed, n = 14) 
#>                  0.5295625 
#> 
#> [[139]]
#> lag(new_confirmed, n = 14) 
#>                  0.8784718 
#> 
#> [[140]]
#> lag(new_confirmed, n = 14) 
#>                  0.4523111 
#> 
#> [[141]]
#> lag(new_confirmed, n = 14) 
#>                 0.06500141 
#> 
#> [[142]]
#> lag(new_confirmed, n = 14) 
#>                  0.7416119 
#> 
#> [[143]]
#> lag(new_confirmed, n = 14) 
#>                  0.8947562 
#> 
#> [[144]]
#> lag(new_confirmed, n = 14) 
#>                  0.7780699 
#> 
#> [[145]]
#> lag(new_confirmed, n = 14) 
#>               -0.003802281 
#> 
#> [[146]]
#> lag(new_confirmed, n = 14) 
#>                   1.044826 
#> 
#> [[147]]
#> lag(new_confirmed, n = 14) 
#>                  0.9334443 
#> 
#> [[148]]
#> lag(new_confirmed, n = 14) 
#>                0.005462774 
#> 
#> [[149]]
#> lag(new_confirmed, n = 14) 
#>                  0.7127593 
#> 
#> [[150]]
#> lag(new_confirmed, n = 14) 
#>                  0.6934141 
#> 
#> [[151]]
#> lag(new_confirmed, n = 14) 
#>                  0.4480632 
#> 
#> [[152]]
#> lag(new_confirmed, n = 14) 
#>                  0.7526479 
#> 
#> [[153]]
#> lag(new_confirmed, n = 14) 
#>                  0.6662816 
#> 
#> [[154]]
#> lag(new_confirmed, n = 14) 
#>                  0.2028236 
#> 
#> [[155]]
#> [1] NA
#> 
#> [[156]]
#> lag(new_confirmed, n = 14) 
#>                  0.9219172 
#> 
#> [[157]]
#> [1] NA
#> 
#> [[158]]
#> lag(new_confirmed, n = 14) 
#>                  0.5813632 
#> 
#> [[159]]
#> [1] NA
#> 
#> [[160]]
#> lag(new_confirmed, n = 14) 
#>                  0.7849124 
#> 
#> [[161]]
#> lag(new_confirmed, n = 14) 
#>                  0.0817926 
#> 
#> [[162]]
#> lag(new_confirmed, n = 14) 
#>                  0.7567328 
#> 
#> [[163]]
#> lag(new_confirmed, n = 14) 
#>                  0.4486768 
#> 
#> [[164]]
#> lag(new_confirmed, n = 14) 
#>                  0.9711064 
#> 
#> [[165]]
#> lag(new_confirmed, n = 14) 
#>                  0.8404479 
#> 
#> [[166]]
#> lag(new_confirmed, n = 14) 
#>                  0.8340365 
#> 
#> [[167]]
#> [1] NA
#> 
#> [[168]]
#> lag(new_confirmed, n = 14) 
#>                   1.146428 
#> 
#> [[169]]
#> [1] NA
#> 
#> [[170]]
#> [1] NA
#> 
#> [[171]]
#> [1] NA
#> 
#> [[172]]
#> [1] NA
#> 
#> [[173]]
#> [1] NA
#> 
#> [[174]]
#> lag(new_confirmed, n = 14) 
#>                  0.9711078 
#> 
#> [[175]]
#> lag(new_confirmed, n = 14) 
#>                  0.9645341 
#> 
#> [[176]]
#> [1] NA
#> 
#> [[177]]
#> lag(new_confirmed, n = 14) 
#>                  0.1992464 
#> 
#> [[178]]
#> [1] NA
#> 
#> [[179]]
#> lag(new_confirmed, n = 14) 
#>                 0.05725861 
#> 
#> [[180]]
#> lag(new_confirmed, n = 14) 
#>                  0.5481366 
#> 
#> [[181]]
#> [1] NA
#> 
#> [[182]]
#> lag(new_confirmed, n = 14) 
#>                -0.05410779 
#> 
#> [[183]]
#> lag(new_confirmed, n = 14) 
#>               -0.007462687 
#> 
#> [[184]]
#> lag(new_confirmed, n = 14) 
#>                  0.3680185 
#> 
#> [[185]]
#> lag(new_confirmed, n = 14) 
#>                -0.00138019 
#> 
#> [[186]]
#> lag(new_confirmed, n = 14) 
#>                  0.8943568 
#> 
#> [[187]]
#> lag(new_confirmed, n = 14) 
#>                  0.4699798 
#> 
#> [[188]]
#> lag(new_confirmed, n = 14) 
#>               -0.001714412 
#> 
#> [[189]]
#> lag(new_confirmed, n = 14) 
#>                -0.01177239 
#> 
#> [[190]]
#> lag(new_confirmed, n = 14) 
#>                  0.4459296 
#> 
#> [[191]]
#> lag(new_confirmed, n = 14) 
#>                   0.741075 
#> 
#> [[192]]
#> lag(new_confirmed, n = 14) 
#>                  0.7394529 
#> 
#> [[193]]
#> lag(new_confirmed, n = 14) 
#>                  0.9754129 
#> 
#> [[194]]
#> lag(new_confirmed, n = 14) 
#>               -0.004582951 
#> 
#> [[195]]
#> lag(new_confirmed, n = 14) 
#>                  0.1134682 
#> 
#> [[196]]
#> lag(new_confirmed, n = 14) 
#>                  0.8579614 
#> 
#> [[197]]
#> lag(new_confirmed, n = 14) 
#>                  0.7022731 
#> 
#> [[198]]
#> lag(new_confirmed, n = 14) 
#>               0.0003349323 
#> 
#> [[199]]
#> lag(new_confirmed, n = 14) 
#>                -0.01591102 
#> 
#> [[200]]
#> lag(new_confirmed, n = 14) 
#>                   0.902218 
#> 
#> [[201]]
#> [1] NA
#> 
#> [[202]]
#> lag(new_confirmed, n = 14) 
#>                  0.2367338 
#> 
#> [[203]]
#> lag(new_confirmed, n = 14) 
#>                  0.6937724 
#> 
#> [[204]]
#> lag(new_confirmed, n = 14) 
#>               8.226172e-05 
#> 
#> [[205]]
#> lag(new_confirmed, n = 14) 
#>                  0.7049068 
#> 
#> [[206]]
#> lag(new_confirmed, n = 14) 
#>                  0.4773192 
#> 
#> [[207]]
#> [1] NA
#> 
#> [[208]]
#> [1] NA
#> 
#> [[209]]
#> lag(new_confirmed, n = 14) 
#>                  0.2683079 
#> 
#> [[210]]
#> lag(new_confirmed, n = 14) 
#>                  0.3773631 
#> 
#> [[211]]
#> lag(new_confirmed, n = 14) 
#>                 0.09471127 
#> 
#> [[212]]
#> lag(new_confirmed, n = 14) 
#>                   0.791494 
#> 
#> [[213]]
#> [1] NA
#> 
#> [[214]]
#> [1] NA
#> 
#> [[215]]
#> lag(new_confirmed, n = 14) 
#>                  0.9127807 
#> 
#> [[216]]
#> lag(new_confirmed, n = 14) 
#>                  0.6683978 
#> 
#> [[217]]
#> lag(new_confirmed, n = 14) 
#>                  0.4777528 
#> 
#> [[218]]
#> lag(new_confirmed, n = 14) 
#>                   0.289256 
#> 
#> [[219]]
#> lag(new_confirmed, n = 14) 
#>                 0.09288966 
#> 
#> [[220]]
#> lag(new_confirmed, n = 14) 
#>                  0.1472352 
#> 
#> [[221]]
#> [1] NA
#> 
#> [[222]]
#> lag(new_confirmed, n = 14) 
#>                  0.8942314 
#> 
#> [[223]]
#> lag(new_confirmed, n = 14) 
#>                  0.8124493 
#> 
#> [[224]]
#> lag(new_confirmed, n = 14) 
#>                0.001282686 
#> 
#> [[225]]
#> lag(new_confirmed, n = 14) 
#>                   1.164049 
#> 
#> [[226]]
#> lag(new_confirmed, n = 14) 
#>                 -0.7243288 
#> 
#> [[227]]
#> lag(new_confirmed, n = 14) 
#>                  0.9300546 
#> 
#> [[228]]
#> lag(new_confirmed, n = 14) 
#>                         NA 
#> 
#> [[229]]
#> [1] NA
#> 
#> [[230]]
#> lag(new_confirmed, n = 14) 
#>                  0.9314963 
#> 
#> [[231]]
#> [1] NA
#> 
#> [[232]]
#> lag(new_confirmed, n = 14) 
#>                  0.3004533 
#> 
#> [[233]]
#> lag(new_confirmed, n = 14) 
#>                  0.9587314 
#> 
#> [[234]]
#> lag(new_confirmed, n = 14) 
#>                  0.2780126 
#> 
#> [[235]]
#> lag(new_confirmed, n = 14) 
#>                  0.4839777 
#> 
#> [[236]]
#> lag(new_confirmed, n = 14) 
#>                  0.4202595
covid19_lstcol$data %>% 
  map(safely_extract_slope) %>% 
  map_dbl("result")

#>   [1]  6.920523e-01  6.435756e-01
#>   [3]  5.329582e-01  5.595175e-01
#>   [5]  5.196389e-01  7.375993e-01
#>   [7]  8.669078e-01  8.971555e-01
#>   [9]            NA  5.299203e-01
#>  [11]  9.182422e-01  9.642226e-01
#>  [13]  2.903094e-01            NA
#>  [15]  8.760548e-01  8.306039e-01
#>  [17]  1.478383e-01  1.038319e+00
#>  [19] -1.417372e-01  3.760244e-01
#>  [21]  2.665434e-01  7.554900e-02
#>  [23]  6.108023e-01  6.446241e-01
#>  [25]  4.912720e-01  6.540097e-01
#>  [27]  4.138478e-01  5.732950e-01
#>  [29]  6.704973e-01  8.998348e-01
#>  [31]  5.736442e-02  5.809416e-01
#>  [33]  1.256359e-01  1.167159e-01
#>  [35]  9.418862e-01            NA
#>  [37]            NA  2.580678e-01
#>  [39]  4.058524e-01            NA
#>  [41]  4.110214e-01  8.410514e-01
#>  [43]  6.855367e-01  1.374103e-01
#>  [45]  2.689623e-01  5.826228e-01
#>  [47]  1.004124e+00  6.459742e-01
#>  [49]  8.822818e-01            NA
#>  [51]  7.600437e-01            NA
#>  [53] -1.675946e-02            NA
#>  [55]  9.635528e-01  1.049798e+00
#>  [57]            NA  5.009491e-01
#>  [59]  1.780097e-01  6.290575e-01
#>  [61]            NA  5.348175e-02
#>  [63]  2.492007e-01  5.076202e-01
#>  [65] -1.289268e-02  1.628999e-01
#>  [67]  8.211096e-01  6.890107e-01
#>  [69]  4.747831e-01            NA
#>  [71]  1.370787e-01  8.498793e-01
#>  [73]  2.613858e-02            NA
#>  [75]  4.039257e-01  2.783185e-01
#>  [77]            NA  8.992036e-01
#>  [79]  7.484860e-01  4.874482e-01
#>  [81]            NA  9.351811e-01
#>  [83]            NA  2.809673e-02
#>  [85]            NA            NA
#>  [87]  8.144507e-01            NA
#>  [89]  2.512326e-01  1.763243e-02
#>  [91]  5.299206e-01  4.224714e-01
#>  [93] -1.196319e-02  2.954188e-01
#>  [95]            NA            NA
#>  [97]  4.821869e-01  7.901768e-01
#>  [99]  8.979504e-01  9.487975e-01
#> [101]  7.877892e-01            NA
#> [103]  8.682548e-01  2.087615e-01
#> [105]  8.144906e-01  7.282967e-01
#> [107]  5.431636e-01  8.678327e-01
#> [109]            NA  1.177228e+00
#> [111]  1.624997e-01  7.411408e-01
#> [113]  2.179744e-02  9.123329e-01
#> [115]  1.588501e-01  2.238890e-02
#> [117]  9.392607e-01  7.137065e-01
#> [119]  1.059572e-01  1.280151e-01
#> [121]  7.095778e-01  8.902059e-01
#> [123]  8.601565e-01  9.333330e-01
#> [125]            NA            NA
#> [127]  7.684239e-01  2.407153e-01
#> [129]  8.741184e-01  8.370697e-01
#> [131]  4.458898e-01  8.722886e-01
#> [133] -1.612903e-02            NA
#> [135]  7.781136e-01  4.789353e-01
#> [137]            NA  5.295625e-01
#> [139]  8.784718e-01  4.523111e-01
#> [141]  6.500141e-02  7.416119e-01
#> [143]  8.947562e-01  7.780699e-01
#> [145] -3.802281e-03  1.044826e+00
#> [147]  9.334443e-01  5.462774e-03
#> [149]  7.127593e-01  6.934141e-01
#> [151]  4.480632e-01  7.526479e-01
#> [153]  6.662816e-01  2.028236e-01
#> [155]            NA  9.219172e-01
#> [157]            NA  5.813632e-01
#> [159]            NA  7.849124e-01
#> [161]  8.179260e-02  7.567328e-01
#> [163]  4.486768e-01  9.711064e-01
#> [165]  8.404479e-01  8.340365e-01
#> [167]            NA  1.146428e+00
#> [169]            NA            NA
#> [171]            NA            NA
#> [173]            NA  9.711078e-01
#> [175]  9.645341e-01            NA
#> [177]  1.992464e-01            NA
#> [179]  5.725861e-02  5.481366e-01
#> [181]            NA -5.410779e-02
#> [183] -7.462687e-03  3.680185e-01
#> [185] -1.380190e-03  8.943568e-01
#> [187]  4.699798e-01 -1.714412e-03
#> [189] -1.177239e-02  4.459296e-01
#> [191]  7.410750e-01  7.394529e-01
#> [193]  9.754129e-01 -4.582951e-03
#> [195]  1.134682e-01  8.579614e-01
#> [197]  7.022731e-01  3.349323e-04
#> [199] -1.591102e-02  9.022180e-01
#> [201]            NA  2.367338e-01
#> [203]  6.937724e-01  8.226172e-05
#> [205]  7.049068e-01  4.773192e-01
#> [207]            NA            NA
#> [209]  2.683079e-01  3.773631e-01
#> [211]  9.471127e-02  7.914940e-01
#> [213]            NA            NA
#> [215]  9.127807e-01  6.683978e-01
#> [217]  4.777528e-01  2.892560e-01
#> [219]  9.288966e-02  1.472352e-01
#> [221]            NA  8.942314e-01
#> [223]  8.124493e-01  1.282686e-03
#> [225]  1.164049e+00 -7.243288e-01
#> [227]  9.300546e-01            NA
#> [229]            NA  9.314963e-01
#> [231]            NA  3.004533e-01
#> [233]  9.587314e-01  2.780126e-01
#> [235]  4.839777e-01  4.202595e-01
32 / 38

Alternatively, `mutate()` with `map()` inside a tibble

covid19_lstcol %>% 
  mutate(
    slopes = map(data, safely_extract_slope) %>% 
      map_dbl("result"))

#> # A tibble: 236 x 3
#>   country_region      data                   slopes
#>   <chr>               <list>                  <dbl>
#> 1 Afghanistan         <tibble[,7] [306 × 7]>  0.692
#> 2 Albania             <tibble[,7] [298 × 7]>  0.644
#> 3 Algeria             <tibble[,7] [306 × 7]>  0.533
#> 4 Andorra             <tibble[,7] [305 × 7]>  0.560
#> 5 Angola              <tibble[,7] [287 × 7]>  0.520
#> 6 Antigua and Barbuda <tibble[,7] [294 × 7]>  0.738
#> # … with 230 more rows

33 / 38

Side-effects

Most functions are called for the value that they return. But some functions are called primarily for their side-effects, for example:

print() displays values/plots to the screen.
write_csv() writes values to the CSV file saved in the disk.

No values/outputs to be captured.

We're going to save each region's data into separate CSV files to make it easier to import in the future.

34 / 38

`group_split()` splits tibble into chunks

covid19_lst <- covid19_byregion %>% 
  group_by(country_region) %>% 
  group_split()
covid19_lst[[1]]

#> # A tibble: 306 x 8
#>   country_region date       confirmed deaths recovered
#>   <chr>          <date>         <dbl>  <dbl>     <dbl>
#> 1 Afghanistan    2020-03-01         1      0         0
#> 2 Afghanistan    2020-03-02         1      0         0
#> 3 Afghanistan    2020-03-03         2      0         0
#> 4 Afghanistan    2020-03-04         4      0         0
#> 5 Afghanistan    2020-03-05         4      0         0
#> 6 Afghanistan    2020-03-06         4      0         0
#> # … with 300 more rows, and 3 more variables:
#> #   new_confirmed <dbl>, new_deaths <dbl>,
#> #   new_recovered <dbl>

35 / 38

`walk()` with side-effects, if no outputs

region_names <- unique(covid19_byregion$country_region)
file_name <- glue::glue("data/covid19/{region_names}.csv")
# fs::dir_create("data/covid19")
walk2(covid19_lst, file_name, write_csv)

fs::dir_ls("data/covid19")

#> data/covid19/Afghanistan.csv
#> data/covid19/Albania.csv
#> data/covid19/Algeria.csv
#> data/covid19/Andorra.csv
#> data/covid19/Angola.csv
#> data/covid19/Antigua and Barbuda.csv
#> data/covid19/Argentina.csv
#> data/covid19/Armenia.csv
#> data/covid19/Aruba.csv
#> data/covid19/Australia.csv
#> data/covid19/Austria.csv
#> data/covid19/Azerbaijan.csv
#> data/covid19/Bahamas, The.csv
#> data/covid19/Bahamas.csv
#> data/covid19/Bahrain.csv
#> data/covid19/Bangladesh.csv
#> data/covid19/Barbados.csv
#> data/covid19/Belarus.csv
#> data/covid19/Belgium.csv
#> data/covid19/Belize.csv
#> data/covid19/Benin.csv
#> data/covid19/Bhutan.csv
#> data/covid19/Bolivia.csv
#> data/covid19/Bosnia and Herzegovina.csv
#> data/covid19/Botswana.csv
#> data/covid19/Brazil.csv
#> data/covid19/Brunei.csv
#> data/covid19/Bulgaria.csv
#> data/covid19/Burkina Faso.csv
#> data/covid19/Burma.csv
#> data/covid19/Burundi.csv
#> data/covid19/Cabo Verde.csv
#> data/covid19/Cambodia.csv
#> data/covid19/Cameroon.csv
#> data/covid19/Canada.csv
#> data/covid19/Cape Verde.csv
#> data/covid19/Cayman Islands.csv
#> data/covid19/Central African Republic.csv
#> data/covid19/Chad.csv
#> data/covid19/Channel Islands.csv
#> data/covid19/Chile.csv
#> data/covid19/China.csv
#> data/covid19/Colombia.csv
#> data/covid19/Comoros.csv
#> data/covid19/Congo (Brazzaville).csv
#> data/covid19/Congo (Kinshasa).csv
#> data/covid19/Costa Rica.csv
#> data/covid19/Cote d'Ivoire.csv
#> data/covid19/Croatia.csv
#> data/covid19/Cruise Ship.csv
#> data/covid19/Cuba.csv
#> data/covid19/Curacao.csv
#> data/covid19/Cyprus.csv
#> data/covid19/Czech Republic.csv
#> data/covid19/Czechia.csv
#> data/covid19/Denmark.csv
#> data/covid19/Diamond Princess.csv
#> data/covid19/Djibouti.csv
#> data/covid19/Dominica.csv
#> data/covid19/Dominican Republic.csv
#> data/covid19/East Timor.csv
#> data/covid19/Ecuador.csv
#> data/covid19/Egypt.csv
#> data/covid19/El Salvador.csv
#> data/covid19/Equatorial Guinea.csv
#> data/covid19/Eritrea.csv
#> data/covid19/Estonia.csv
#> data/covid19/Eswatini.csv
#> data/covid19/Ethiopia.csv
#> data/covid19/Faroe Islands.csv
#> data/covid19/Fiji.csv
#> data/covid19/Finland.csv
#> data/covid19/France.csv
#> data/covid19/French Guiana.csv
#> data/covid19/Gabon.csv
#> data/covid19/Gambia, The.csv
#> data/covid19/Gambia.csv
#> data/covid19/Georgia.csv
#> data/covid19/Germany.csv
#> data/covid19/Ghana.csv
#> data/covid19/Gibraltar.csv
#> data/covid19/Greece.csv
#> data/covid19/Greenland.csv
#> data/covid19/Grenada.csv
#> data/covid19/Guadeloupe.csv
#> data/covid19/Guam.csv
#> data/covid19/Guatemala.csv
#> data/covid19/Guernsey.csv
#> data/covid19/Guinea-Bissau.csv
#> data/covid19/Guinea.csv
#> data/covid19/Guyana.csv
#> data/covid19/Haiti.csv
#> data/covid19/Holy See.csv
#> data/covid19/Honduras.csv
#> data/covid19/Hong Kong SAR.csv
#> data/covid19/Hong Kong.csv
#> data/covid19/Hungary.csv
#> data/covid19/Iceland.csv
#> data/covid19/India.csv
#> data/covid19/Indonesia.csv
#> data/covid19/Iran (Islamic Republic of).csv
#> data/covid19/Iran.csv
#> data/covid19/Iraq.csv
#> data/covid19/Ireland.csv
#> data/covid19/Israel.csv
#> data/covid19/Italy.csv
#> data/covid19/Jamaica.csv
#> data/covid19/Japan.csv
#> data/covid19/Jersey.csv
#> data/covid19/Jordan.csv
#> data/covid19/Kazakhstan.csv
#> data/covid19/Kenya.csv
#> data/covid19/Kosovo.csv
#> data/covid19/Kuwait.csv
#> data/covid19/Kyrgyzstan.csv
#> data/covid19/Laos.csv
#> data/covid19/Latvia.csv
#> data/covid19/Lebanon.csv
#> data/covid19/Lesotho.csv
#> data/covid19/Liberia.csv
#> data/covid19/Libya.csv
#> data/covid19/Liechtenstein.csv
#> data/covid19/Lithuania.csv
#> data/covid19/Luxembourg.csv
#> data/covid19/MS Zaandam.csv
#> data/covid19/Macao SAR.csv
#> data/covid19/Macau.csv
#> data/covid19/Madagascar.csv
#> data/covid19/Malawi.csv
#> data/covid19/Malaysia.csv
#> data/covid19/Maldives.csv
#> data/covid19/Mali.csv
#> data/covid19/Malta.csv
#> data/covid19/Marshall Islands.csv
#> data/covid19/Martinique.csv
#> data/covid19/Mauritania.csv
#> data/covid19/Mauritius.csv
#> data/covid19/Mayotte.csv
#> data/covid19/Mexico.csv
#> data/covid19/Moldova.csv
#> data/covid19/Monaco.csv
#> data/covid19/Mongolia.csv
#> data/covid19/Montenegro.csv
#> data/covid19/Morocco.csv
#> data/covid19/Mozambique.csv
#> data/covid19/Namibia.csv
#> data/covid19/Nepal.csv
#> data/covid19/Netherlands.csv
#> data/covid19/New Zealand.csv
#> data/covid19/Nicaragua.csv
#> data/covid19/Niger.csv
#> data/covid19/Nigeria.csv
#> data/covid19/North Macedonia.csv
#> data/covid19/Norway.csv
#> data/covid19/Oman.csv
#> data/covid19/Others.csv
#> data/covid19/Pakistan.csv
#> data/covid19/Palestine.csv
#> data/covid19/Panama.csv
#> data/covid19/Papua New Guinea.csv
#> data/covid19/Paraguay.csv
#> data/covid19/Peru.csv
#> data/covid19/Philippines.csv
#> data/covid19/Poland.csv
#> data/covid19/Portugal.csv
#> data/covid19/Puerto Rico.csv
#> data/covid19/Qatar.csv
#> data/covid19/Republic of Ireland.csv
#> data/covid19/Republic of Korea.csv
#> data/covid19/Republic of Moldova.csv
#> data/covid19/Republic of the Congo.csv
#> data/covid19/Reunion.csv
#> data/covid19/Romania.csv
#> data/covid19/Russia.csv
#> data/covid19/Russian Federation.csv
#> data/covid19/Rwanda.csv
#> data/covid19/Saint Barthelemy.csv
#> data/covid19/Saint Kitts and Nevis.csv
#> data/covid19/Saint Lucia.csv
#> data/covid19/Saint Martin.csv
#> data/covid19/Saint Vincent and the Grenadines.csv
#> data/covid19/Samoa.csv
#> data/covid19/San Marino.csv
#> data/covid19/Sao Tome and Principe.csv
#> data/covid19/Saudi Arabia.csv
#> data/covid19/Senegal.csv
#> data/covid19/Serbia.csv
#> data/covid19/Seychelles.csv
#> data/covid19/Sierra Leone.csv
#> data/covid19/Singapore.csv
#> data/covid19/Slovakia.csv
#> data/covid19/Slovenia.csv
#> data/covid19/Solomon Islands.csv
#> data/covid19/Somalia.csv
#> data/covid19/South Africa.csv
#> data/covid19/South Korea.csv
#> data/covid19/South Sudan.csv
#> data/covid19/Spain.csv
#> data/covid19/Sri Lanka.csv
#> data/covid19/St. Martin.csv
#> data/covid19/Sudan.csv
#> data/covid19/Suriname.csv
#> data/covid19/Sweden.csv
#> data/covid19/Switzerland.csv
#> data/covid19/Syria.csv
#> data/covid19/Taipei and environs.csv
#> data/covid19/Taiwan*.csv
#> data/covid19/Taiwan.csv
#> data/covid19/Tajikistan.csv
#> data/covid19/Tanzania.csv
#> data/covid19/Thailand.csv
#> data/covid19/The Bahamas.csv
#> data/covid19/The Gambia.csv
#> data/covid19/Timor-Leste.csv
#> data/covid19/Togo.csv
#> data/covid19/Trinidad and Tobago.csv
#> data/covid19/Tunisia.csv
#> data/covid19/Turkey.csv
#> data/covid19/UK.csv
#> data/covid19/US.csv
#> data/covid19/Uganda.csv
#> data/covid19/Ukraine.csv
#> data/covid19/United Arab Emirates.csv
#> data/covid19/United Kingdom.csv
#> data/covid19/Uruguay.csv
#> data/covid19/Uzbekistan.csv
#> data/covid19/Vanuatu.csv
#> data/covid19/Vatican City.csv
#> data/covid19/Venezuela.csv
#> data/covid19/Viet Nam.csv
#> data/covid19/Vietnam.csv
#> data/covid19/West Bank and Gaza.csv
#> data/covid19/Yemen.csv
#> data/covid19/Zambia.csv
#> data/covid19/Zimbabwe.csv
#> data/covid19/occupied Palestinian territory.csv

36 / 38

{fs} provides a cross-platform, uniform interface to file system operations.

library(fs)
covid19_local <- map_dfr(
  dir_ls("data/covid19", glob = "*.csv"), read_csv)
covid19_local

#> # A tibble: 56,107 x 8
#>   country_region date       confirmed deaths recovered
#>   <chr>          <date>         <dbl>  <dbl>     <dbl>
#> 1 Afghanistan    2020-03-01         1      0         0
#> 2 Afghanistan    2020-03-02         1      0         0
#> 3 Afghanistan    2020-03-03         2      0         0
#> 4 Afghanistan    2020-03-04         4      0         0
#> 5 Afghanistan    2020-03-05         4      0         0
#> 6 Afghanistan    2020-03-06         4      0         0
#> # … with 56,101 more rows, and 3 more variables:
#> #   new_confirmed <dbl>, new_deaths <dbl>,
#> #   new_recovered <dbl>

37 / 38

Reading

Functionals

38 / 38

Help

Keyboard shortcuts

↑, ←, Pg Up, k

Go to previous slide

↓, →, Pg Dn, Space, j

Go to next slide

Home

Go to first slide

End

Go to last slide

Number + Return

Go to specific slide

b / m / f

Toggle blackout / mirrored / fullscreen mode

Clone slideshow

Toggle presenter mode

Restart the presentation timer

?, h

Toggle this help

STATS 220

Don’t repeat yourself 🐈

Case study: batch download and process COVID-19 data

Keep calm, and import CSVs one by one

Data is changing

What are the repeated patterns?

DRY: automate the common tasks above

1 generate a sequence of urls

Test make_url(): does it return what we expected?

2 & 3 read data and standardise column names

Why function

The joy of functional programming (FP)

comes to rescue

copy and paste

- map()

- map()

- map()

- map()

- map()

- map()

- map()

Bind multiple tibbles by row

- map()

Hmm, where are the dates?

- map()

- map2()

- map()

- map2()

- map()

- map2()

- map_dfr()

- map()

- map2()

- map_dfr()

- map_*()

Wrangle a bit

Apply a function across() multiple columns

Look inside the list-column

Fit regression models and extract slopes

Function time

Pull out results

Alternatively, mutate() with map() inside a tibble

Side-effects

group_split() splits tibble into chunks

walk() with side-effects, if no outputs

Reading

Help

Test `make_url()`: does it return what we expected?

- `map()`

- `map()`

- `map()`

- `map()`

- `map()`

- `map()`

- `map()`

- `map()`

- `map()`

- `map2()`

- `map()`

- `map2()`

- `map()`

- `map2()`

- `map_dfr()`

- `map()`

- `map2()`

- `map_dfr()`

- `map_*()`

Apply a function `across()` multiple columns

Alternatively, `mutate()` with `map()` inside a tibble

`group_split()` splits tibble into chunks

`walk()` with side-effects, if no outputs