DBnomics

What You're Getting

DBnomics is a meta-aggregator that mirrors data from 80+ statistical providers into a single unified API. Instead of building integrations to each source individually, you query one API with a provider/dataset/series pattern. DBnomics does not transform the data — it mirrors as-is. Particularly useful for cross-country comparisons pulling from multiple sources in one pipeline.

Ingestion Strategy

# pip install dbnomics
import dbnomics

# Single series: IMF World Economic Outlook, Germany GDP growth
df = dbnomics.fetch_series("IMF", "WEO:2024-10", "DEU.NGDP_RPCH")
print(df[["period", "value"]].tail(10))

# Multiple series across providers in one call
series_list = [
    "FRED/UNRATE",
    "Eurostat/une_rt_m/M.SA.TOTAL.PC_ACT.T.DE",
    "BLS/pr/PRS85006032",
]
df_multi = dbnomics.fetch_series(series_list)

# Direct REST API
import requests
r = requests.get(
    "https://api.db.nomics.world/v22/series",
    params={"provider_code": "FRED", "dataset_code": "UNRATE", "limit": 1000}
)
data = r.json()["series"]["docs"]

Rate Limits & API Details

• No published rate limit; reasonable use expected
• REST API: https://api.db.nomics.world/v22
• Endpoints: /series, /datasets, /providers
• Python: pip install dbnomics
• R: rdbnomics on CRAN
• Catalog browser: db.nomics.world

Schema Stability

API v22 schema is stable. Underlying series codes mirror each provider — when a provider changes codes, DBnomics mirrors the change. Check last_fetch_at on each series to see when DBnomics last synced from source.

Data Quality Gotchas

• Data lag: mirrors on its own schedule — may lag primary source by hours to days
• Partial coverage: not all series from every provider are indexed
• Series codes: Eurostat and OECD codes encode dimension selections — use the catalog browser to find exact codes
• Null handling: missing values appear as null in JSON; Python library fills with NaN
• Units not normalized: period formats differ across providers — normalize dates before joining