Compare DfT traffic count data with DfT coronavirus stats for cycling

[Robinlovelace]

Will be an interesting comparison. Here's the DfT coronavirus data:

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
library(ggplot2)

u = "https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/942559/COVID-19-transport-use-statistics.ods"
f = basename(u)
download.file(u, f)
d_original = tibble::tibble(readODS::read_ods(f, skip = 6))
names(d_original) = gsub(pattern = "\\d", replacement = "", x = names(d_original))
d = janitor::clean_names(d_original)
d = d[nchar(d[[1]]) == 10, ]
d$date = lubridate::dmy(d$date_weekends_and_bank_holidays_in_grey)
d = d %>% select(-(cars:heavy_goods_vehicles))

d = d %>%
  mutate(across(all_motor_vehicles:cycling, as.numeric))
#> Warning: Problem with `mutate()` input `..1`.
#> ℹ NAs introduced by coercion
#> ℹ Input `..1` is `across(all_motor_vehicles:cycling, as.numeric)`.
#> Warning in fn(col, ...): NAs introduced by coercion
#> Warning: Problem with `mutate()` input `..1`.
#> ℹ NAs introduced by coercion
#> ℹ Input `..1` is `across(all_motor_vehicles:cycling, as.numeric)`.
#> Warning in fn(col, ...): NAs introduced by coercion
#> Warning: Problem with `mutate()` input `..1`.
#> ℹ NAs introduced by coercion
#> ℹ Input `..1` is `across(all_motor_vehicles:cycling, as.numeric)`.
#> Warning in fn(col, ...): NAs introduced by coercion
#> Warning: Problem with `mutate()` input `..1`.
#> ℹ NAs introduced by coercion
#> ℹ Input `..1` is `across(all_motor_vehicles:cycling, as.numeric)`.
#> Warning in fn(col, ...): NAs introduced by coercion
  # mutate(across(cars:cycling, as.numeric))
# summary(d)

droll = d %>%
  mutate(across(all_motor_vehicles:cycling, zoo::rollmean, k = 30, align = "right", fill = NA))
  # mutate(across(cars:cycling, zoo::rollmean, k = 30, align = "right", fill = NA))
# d
# nrow(d)
d_long = tidyr::pivot_longer(d, cols = all_motor_vehicles:cycling)
droll_long = tidyr::pivot_longer(droll, cols = all_motor_vehicles:cycling)
d_long$date = as.Date(d_long$date)
droll_long$date = as.Date(droll_long$date)
g = ggplot(droll_long) +
  geom_line(aes(date, value, colour = name), size = 1.3) +
  geom_line(aes(date, value, colour = name), alpha = 0.3, data = d_long) +
  scale_y_continuous(labels = scales::percent) +
  theme_bw()
g
#> Warning: Removed 233 row(s) containing missing values (geom_path).
#> Warning: Removed 51 row(s) containing missing values (geom_path).

^{Created on 2020-12-17 by the reprex package (v0.3.0)}