Get started with tidyBdE

tidyBdE is an R package that retrieves time series data from Banco de España bulk CSV files and the Statistics web service (API). Data are returned as tibble objects. The package infers date, character and numeric column types where possible. Bulk CSV helpers identify series with stable sequential numbers (Numero_secuencial), while API helpers use Nombre_de_la_serie as the API series code.

Search for time series

Banco de España (BdE) publishes numerous time series produced by the institution or compiled from other sources, such as Eurostat or INE.

The basic entry point for discovering time series is catalog metadata. You can search for time series by name:

library(tidyBdE)

library(ggplot2)
library(dplyr)
library(tidyr)

# Search for GBP in the "TC" (exchange rate) catalog metadata.
xr_gbp <- bde_catalog_search("GBP", catalog = "TC")

xr_gbp |>
  select(Numero_secuencial, Descripcion_de_la_serie) |>
  # Display the table in the document.
  knitr::kable()

Note: BdE catalog metadata is currently available in Spanish only, so search terms must be in Spanish to retrieve results. Banco de España is working on an English version.

After finding a time series, load the GBP/EUR exchange rate from bulk CSV files using its stable sequential number (Numero_secuencial):

seq_number <- xr_gbp |>
  # Select the first record.
  slice(1) |>
  # Get the stable sequential number.
  pull(Numero_secuencial) |>
  # Convert to numeric.
  as.double()

seq_number
#> [1] 573214

time_series <- bde_series_load(seq_number, series_label = "EUR_GBP_XR") |>
  filter(Date >= "2010-01-01" & Date <= "2020-12-31") |>
  drop_na()

time_series
#> # A tibble: 2,816 × 2
#>    Date       EUR_GBP_XR
#>    <date>          <dbl>
#>  1 2010-01-04      0.891
#>  2 2010-01-05      0.900
#>  3 2010-01-06      0.899
#>  4 2010-01-07      0.900
#>  5 2010-01-08      0.893
#>  6 2010-01-11      0.899
#>  7 2010-01-12      0.897
#>  8 2010-01-13      0.895
#>  9 2010-01-14      0.890
#> 10 2010-01-15      0.881
#> # ℹ 2,806 more rows

Plot time series

The package also provides a custom ggplot2 theme based on BdE publications:

ggplot(time_series, aes(x = Date, y = EUR_GBP_XR)) +
  geom_line(colour = bde_tidy_palettes(n = 1)) +
  geom_smooth(method = "gam", colour = bde_tidy_palettes(n = 2)[2]) +
  labs(
    title = "EUR/GBP exchange rate (2010-2020)",
    subtitle = "%",
    caption = "Source: BdE"
  ) +
  geom_vline(
    xintercept = as.Date("2016-06-23"),
    linetype = "dotted"
  ) +
  geom_label(aes(
    x = as.Date("2016-06-23"),
    y = 0.95,
    label = "Brexit"
  )) +
  coord_cartesian(ylim = c(0.7, 1)) +
  theme_tidybde()
Figure 1: EUR/GBP exchange rate (2010-2020)

Figure 1: EUR/GBP exchange rate (2010-2020)

The package also provides convenience functions for selected Spanish macroeconomic indicators, so you do not need to search for them manually:

# Data in long format.

plotseries <- bde_ind_gdp_var("GDP YoY", out_format = "long") |>
  bind_rows(
    bde_ind_unemployment_rate("Unemployment Rate", out_format = "long")
  ) |>
  drop_na() |>
  filter(Date >= "2010-01-01" & Date <= "2019-12-31")

ggplot(plotseries, aes(x = Date, y = serie_value)) +
  geom_line(aes(color = serie_name), linewidth = 1) +
  labs(
    title = "Spanish economic indicators (2010-2019)",
    subtitle = "%",
    caption = "Source: BdE"
  ) +
  theme_tidybde() +
  scale_color_bde_d(palette = "bde_vivid_pal") # Use a custom package palette.
Figure 2: Spanish economic indicators (2010-2019)

Figure 2: Spanish economic indicators (2010-2019)

A note on caching

Create a local cache by setting the following option:

options(bde_cache_dir = "./path/to/location")

When this option is set, tidyBdE looks for cached bulk CSV files in the bde_cache_dir directory and loads them to speed up data retrieval.

Update cached data after monthly or quarterly releases with the following commands:

bde_catalog_update()

# Or use `update_cache = TRUE` in most functions.

bde_series_load(573214, update_cache = TRUE)