Singapore Government Data (data.gov.sg) — HDB, COE, Economy → CSV

Download Singapore government statistics as a clean CSV, JSON or Excel file — HDB dwelling units, COE quotas, demography, labour market, education, health — joined and ready for Excel, Google Sheets, or your AI pipeline. No API key, no scraping.

This Actor is a value layer on the official data.gov.sg API (Singapore Open Data Licence, CC BY 4.0-based). It does not scrape the website — it joins, cleans, and packages government open data into one flat table per run.

What you get

Run the default preset (hdb-housing) and the dataset contains real rows like this:

year	hdb_units_managed	hdb_units_completed
2008	883896.0	1769.0
2009	890212.0	7050.0
2010	901971.0	11888.0
2011	916842.0	19005.0
2012	924729.0	12744.0

(14 rows, financial years 2008–2021 — actual output of the hdb-housing preset.)

The rows land in the Dataset tab (exportable as CSV/JSON/Excel with one click) and the file is also stored in the key-value store in your chosen format.

Which preset do you need?

You want…	Use preset
HDB dwelling units under management / completed	hdb-housing
Deaths and live births by year	demography
Female LFPR, employment rate, long-term unemployment	labour-market
COE quotas and cars by engine capacity	vehicles-coe
Primary/secondary enrolment and teachers	education
Hospital admissions, doctors, life expectancy	health
Your own dataset IDs, keys and filters	custom

Legacy codes S1–S6 are still accepted as aliases.

Input

field	type	default	description
mode	enum	studio	studio (presets/joins) or extract (raw pull of 1–5 datasets by ID)
preset	enum	hdb-housing	see table above, or custom
datasetIds	string list	—	extract mode: data.gov.sg dataset ids (d_..., 1–5)
customDatasets	string list	—	custom join: dataset ids (d_..., 2–6)
customKey	string list	["year"]	custom join: key columns (year = auto-extracted)
customJoinType	enum	inner	inner / left
customParams	string	—	JSON object of extra API params (e.g. {"filters": {"sex": "female"}})
customAgg	enum	none	none / sum / mean
format	enum	json	json / csv / xlsx
webhookUrl	string	—	POST the result file here after the run (optional)
aiReady	bool	true	Emit JSON Schema + dictionary + samples
includeAttribution	bool	true	Attach licence metadata sidecar

Output

• Dataset (default) — the actual data rows, one record per row; the final record is the run summary (mode, preset, row count, files, year range).
• Key-value store — <preset>_join.<json|csv|xlsx>, <preset>_ai_ready.json (schema + data dictionary + sample rows for LLM/RAG), <preset>_attribution.json (source metadata), <preset>_verification.json (row count, year range, source rows).

Cost

Pay per joined pull (single joined-pull event). The underlying data comes from Singapore's official free API (data.gov.sg) — you pay for the join + packaging, not the data. Apify compute is billed on top of the event price (paid by the user).

Data & licence

Source: data.gov.sg — Singapore's official open data portal (Singapore Department of Statistics, HDB, MOH, MOE, MOM, LTA, ICA, NEA, and more). Licensed under the Singapore Open Data Licence (CC BY 4.0-based). Attribution is attached to every output.

Rate limits

data.gov.sg's public API is keyless but rate-limited (datastore: 4 calls/10s without a key). The actor paces its fetches within the limit; large datasets take longer but never fail on rate limits.