Argentine public data
Daily, normalized copies of 13 public datasets from 5 Argentine publishers, as CSV, Parquet and a DuckDB database. Updated 2026-11-18 22:09 UTC.
Built with teramot-signals, open source (Apache-2.0). Every dataset keeps its publisher's license: credit the source when you reuse it.
Query everything with DuckDB
ATTACH 'https://data.teramot.com/signals/v1/signals.duckdb' AS signals (READ_ONLY);
USE signals;
SHOW TABLES;
SELECT * FROM bcra__fx_currencies__latest LIMIT 10;
Machine-readable index with checksums: catalog.json.
Datasets
bcra.fx_currencies
BCRA currency master: ISO code (`codigo`) and name (`denominacion`), one row per currency.
CSV Parquet Full history (Parquet)
Columns
codigo, denominacion, source, dataset, extraction_date, extracted_at, provenance, revision_of
bcra.fx_history
Official exchange-rate history per currency (since 2003 with `--since 2003-01-01`). Same variables as `fx_quotes`: `<ISO>.rate` (pesos per unit) and `<ISO>.usd_parity` (parity against the dollar; for gold and silver, USD per ounce).
CSV Parquet Full history (Parquet)
Columns
variable, ref_date, value, source, dataset, extraction_date, extracted_at, provenance, revision_of
bcra.fx_quotes
Official BCRA exchange rates for the last business day, every currency, two variables per currency: `<ISO>.rate` = pesos per unit of the currency; `<ISO>.usd_parity` = parity against the US dollar. Official rate only (no MEP, CCL or parallel "blue" rate).
CSV Parquet Full history (Parquet)
Columns
variable, ref_date, value, source, dataset, extraction_date, extracted_at, provenance, revision_of
bcra.monetary_catalog
Catalog of BCRA monetary variables (one row per `idVariable`): description, category, frequency, unit, currency, first and last reported date and latest value. Columns as published by the BCRA (Spanish). Use it to pick which ids to add to `monetary_series`.
CSV Parquet Full history (Parquet)
Columns
idVariable, descripcion, categoria, tipoSerie, periodicidad, unidadExpresion, moneda, primerFechaInformada, ultFechaInformada, ultValorInformado, source, dataset, extraction_date, extracted_at, provenance, revision_of
bcra.monetary_series
Key BCRA monetary variables, one row per (variable, date). - international_reserves_musd: international reserves, millions of USD, daily. - usd_retail_avg_selling_rate: pesos per dollar, banks' average selling rate, daily. - usd_wholesale_rate_com_a3500: pesos per dollar, Communication A 3500 reference rate, daily. - badlar_private_banks_apr / tamar_private_banks_apr: wholesale deposit rates, % annual nominal. - time_deposit_30d_apr / personal_loans_apr: interest rates, % annual nominal. - monetary_base_mars / total_deposits_mars / private_sector_loans_mars: millions of pesos. - monthly_inflation_pct / yoy_inflation_pct: CPI change (INDEC, republished by the BCRA). - expected_inflation_12m_pct: median 12-month expectation from the REM survey, %. - cer, uva, icl_rent_index: adjustment indexes (CER, UVA purchasing-power unit in pesos, ICL rental contracts index).
CSV Parquet Full history (Parquet)
Columns
variable, ref_date, value, source, dataset, extraction_date, extracted_at, provenance, revision_of
datos_gob_ar.series
Monthly INDEC series republished by datos.gob.ar. Per-variable descriptions come from the API itself (metadata=full) and are appended below.
CSV Parquet Full history (Parquet)
Columns
variable, ref_date, value, source, dataset, extraction_date, extracted_at, provenance, revision_of
datos_gob_ar.sspm
Full history of 108 series from the catalog of the Subsecretaría de Programación Macroeconómica (CPI by region and division, EMAE economic activity and sectors, wages and RIPTE, supermarkets, shopping centers, BNA exchange rate). One row per (variable, date). Units and original source per variable below and in `sspm_catalog`. License CC BY 4.0.
CSV Parquet Full history (Parquet)
Columns
variable, ref_date, value, source, dataset, extraction_date, extracted_at, provenance, revision_of
datos_gob_ar.sspm_catalog
Metadata for the `sspm` series: topic group, the agency's title and description (Spanish), unit, frequency (ISO 8601), first and last date, dataset, primary source, license and source CSV.
CSV Parquet Full history (Parquet)
Columns
series_id, variable, group, title, description, unit, frequency, start_date, end_date, dataset_title, source, publisher, license, distribution_url, dataset, extraction_date, extracted_at, provenance, revision_of
georef.census_localities
INDEC census localities: 8-digit id, name, category (simple, compound, component), role (department seat, capital), department, local government, province and centroid.
CSV Parquet Full history (Parquet)
Columns
categoria, centroide_lat, centroide_lon, departamento_id, departamento_nombre, fuente, funcion, gobierno_local_id, gobierno_local_nombre, id, nombre, provincia_id, provincia_nombre, source, dataset, extraction_date, extracted_at, provenance, revision_of
georef.departments
Departments and partidos (second administrative level): 5-digit INDEC id, name, province and centroid.
CSV Parquet Full history (Parquet)
Columns
categoria, centroide_lat, centroide_lon, fuente, id, nombre, nombre_completo, provincia_id, provincia_interseccion, provincia_nombre, source, dataset, extraction_date, extracted_at, provenance, revision_of
georef.provinces
The 24 jurisdictions (23 provinces and CABA): 2-digit INDEC id, ISO 3166-2 code, name, full name and centroid (lat/lon).
CSV Parquet Full history (Parquet)
Columns
categoria, centroide_lat, centroide_lon, fuente, id, iso_id, iso_nombre, nombre, nombre_completo, source, dataset, extraction_date, extracted_at, provenance, revision_of
indec.supermarkets_by_province
INDEC Supermarket Survey, tables 5 and 6: one row per (month, jurisdiction), since 2017. `jurisdiction` keeps INDEC's labels: «Total del país» (national total), CABA, «24 partidos del Gran Buenos Aires», «Resto de Buenos Aires» and each province. Sales in thousands of current pesos; floor area in m²; sales per m² and per transaction in pesos; stores and transactions in units.
CSV Parquet Full history (Parquet)
Columns
date, jurisdiction, sales_total, stores, sales_per_store, floor_area_m2, sales_per_m2, transactions, sales_per_transaction, sales_total_products, sales_beverages, sales_groceries, sales_bakery, sales_dairy, sales_meat, sales_fruit_vegetables, sales_prepared_food, sales_cleaning_personal_care, sales_clothing_footwear_textiles, sales_electronics_household, sales_other, source, dataset, extraction_date, extracted_at, provenance, revision_of
unr_dataverse.eph_dictionary
EPH dictionary: one row per (variable, value). `code` is INDEC's code («01»), `value_label` its label («GBA») and `original_value` the source text («01 = Gran Buenos Aires»). `methodology` says whether the variable applies before the 2024Q4 questionnaire change (`legacy`), after it (`new`) or always (`both`). Variables without coded values have a null `code`. Labels are in Spanish, as published. Source: Observatorio Económico Social, UNR.
CSV Parquet Full history (Parquet)
Columns
variable, position, indec_variable, type, label, description, base, category, section, methodology, origin, value_position, code, value_label, original_value, source, dataset, extraction_date, extracted_at, provenance, revision_of