Skip to content

Instantly share code, notes, and snippets.

@mschnetzer
Last active September 28, 2023 13:27
Show Gist options
  • Star 2 You must be signed in to star a gist
  • Fork 1 You must be signed in to fork a gist
  • Save mschnetzer/875c0bbf047be3257c258b5011e1ac1d to your computer and use it in GitHub Desktop.
Save mschnetzer/875c0bbf047be3257c258b5011e1ac1d to your computer and use it in GitHub Desktop.
library(tidyverse)
library(eurostat)
library(countrycode)
library(gt)
library(gtExtras)
data <- get_eurostat("prc_hicp_manr", filters = list(coicop = "CP00"))
ea.avg <- data |> filter(geo == "EA") |> slice_max(time, n=1) |> pull(values)
findat <- data |>
left_join(codelist |> select(eurostat, country.name.de), by = c("geo"="eurostat")) |>
filter(time >= "2019-01-01", geo %in% c(eurostat::ea_countries$code, "HR")) |>
select(geo, country.name.de, time, values) |>
drop_na()
trend <- findat |> summarise(trend = list(values), .by = geo)
plotdat <- findat |> slice_max(time, n=1, by = geo) |>
mutate(abw = values - ea.avg) |>
left_join(trend)
plotdat |> arrange(desc(values)) |>
mutate(geo = ifelse(geo == "EL", "GR", geo)) |>
gt() |>
fmt_number(columns = values, locale = "de", decimals = 1, pattern = "{x}%") |>
fmt_number(columns = abw, locale = "de", decimals = 1, force_sign = T) |>
fmt_date(columns = time, rows = everything(), date_style = "yM", locale = "en") |>
fmt_flag(columns = geo, height = "1.5em") |>
gt_plt_sparkline(column = trend, same_limit = F,
palette = c(rep("black", 2), rep("transparent", 3))) |>
cols_label(geo = "", country.name.de = "",
time = "Monat", values = "Inflation", abw = "Abw. Ø") |>
cols_width(country.name.de ~ px(150), values ~ px(100), abw ~ px(80)) |>
cols_align(align = "center", columns = c(time, values, abw)) |>
data_color(columns = values, method = "numeric", palette = "Reds", alpha = 0.9) |>
gt_highlight_rows(rows = geo == "AT", columns = c(geo, country.name.de, time),
target_col = country.name.de, bold_target_only = T, fill = "transparent") |>
tab_header(title = html("Inflationsraten in der Eurozone")) |>
tab_footnote(footnote = "Harmonisierte Verbraucherpreisindizes",
locations = cells_column_labels(values)) |>
tab_footnote(footnote = "Abweichung zum Durchschnitt der Eurozone",
locations = cells_column_labels(abw)) |>
tab_source_note(source_note = html("<p style='text-align:right;'>Daten: Eurostat. Grafik: @matschnetzer</p>")) |>
gt_theme_538() |>
tab_options(heading.title.font.size = 24, footnotes.padding = 0,
footnotes.font.size = 10,
source_notes.font.size = 10) |>
gtsave(filename = "eu_inflation_table.html")
@lbarqueira
Copy link

Hi, thanks for sharing the code, I`m trying to replicate the code in other to learn, and a warning occurs in RStudio:

Warning message:
There was 1 warning in filter().
ℹ In argument: time >= "2019-01-01".
Caused by warning in Ops.factor():
! ‘>=’ not meaningful for factors

Could you please help me how to solve this warning?
Thank you,
Luis Barqueira

@mschnetzer
Copy link
Author

mschnetzer commented May 2, 2023

Hi Luis,

seems like the class of the time variable is factor but should be date. Actually, eurostat::get_eurostat has a default option time_format = "date" but perhaps it did not work in your setup.

Try installing the latest version of the eurostat package (remotes::install_github("ropengov/eurostat")) and reload the data (and clean your cache before eurostat::clean_eurostat_cache()) or otherwise manually change the class to date.

Best, Matthias

@lbarqueira
Copy link

Hi Matthias,

thanks for all your support. After trying all your suggestions the one that solved the problem was to change the class of time variable to date, by using: data$time <- ymd(paste0(as.character(data$time), "-01"))

Everything works fine.

Thanks a lot, looking forward to your next project,
Luis

@mschnetzer
Copy link
Author

Glad to read that it worked out. For more "efficient" coding you could even use the truncated option in ymd to parse incomplete dates:

mutate(time = ymd(time, truncated =2))

Best, MS

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment