China NBS Statistics | 中国国家数据

Search and retrieve China's national statistics from data.stats.gov.cn. This Actor supports dataset searches, monthly/quarterly/annual catalog browsing, indicator metadata, and nationwide statistical time series. It is an unofficial tool, not affiliated with or endorsed by the National Bureau of Statistics of China.

查询中国国家统计局“国家数据”的公开数据：搜索指标和数据集、浏览全国月度/季度/年度目录、获取指标 UUID 和单位、查询全国统计时间序列。本工具非国家统计局官方产品。

Quick start / 快速开始

Run this input to search GDP:

{
  "action": "search",
  "search": "国内生产总值",
  "pageSize": 5,
  "maxResults": 20
}

Open Dataset → Statistics to view results. The Identifiers view shows IDs needed for follow-up queries. JSON export includes each row's original upstream metadata in raw. CSV/Excel exports are available through Apify.

运行后在 Dataset 中查看统计结果；Identifiers 视图提供后续查询所需编号。JSON 导出保留原始元数据；可通过 Apify 导出 CSV 或 Excel。

Operations / 查询方式

action	Purpose	Required inputs
search	Search datasets and indicator observations / 搜索数据	search
catalog	Browse one catalog level / 浏览一层目录	code (1 monthly, 2 quarterly, 3 annual)
indicators	List indicator UUIDs, units, notes / 指标列表	cid
statistics	Retrieve nationwide time series / 全国统计数值	cid, indicatorIds, dts, rootId

Each run performs one operation and one upstream request. There is no automatic pagination, recursive catalog traversal, or automatic choice of a GDP series. Chinese keywords work best. search defaults to page 1 and page size 5 (maximum 10). Upstream may return several indicator rows per dataset, so page size is not necessarily the number of output rows.

每次运行执行一项查询。搜索结果可能同时包含累计值、当季值、增长速度等不同口径，请按指标名称和单位选择。maxResults 限制保存条数，不会自动抓取更多页。

Retrieve statistics / 查询时间序列

1. Run search and copy a dataset cid.
2. Run indicators with that cid, and copy the desired indicatorId.
3. Run catalog with the matching code and no pid; copy the root nodeId as rootId.
4. Run statistics with those IDs and a matching time range.

Catalog example:

{"action":"catalog","code":"2"}

Quarterly GDP indicator example:

{"action":"indicators","cid":"28d936104e304aa191e338eb82b6dc09"}

Example of cumulative GDP for the first two quarters of 2026 (IDs verified on September 9, 2026; obtain fresh IDs if the upstream catalog changes):

{
  "action": "statistics",
  "cid": "28d936104e304aa191e338eb82b6dc09",
  "indicatorIds": ["8c5fab362d124fa7b91af833b3bd7397"],
  "dts": ["202601SS-202602SS"],
  "rootId": "a94b8b7365a94874968cabbe392cf679",
  "maxResults": 100
}

Monthly ranges use MM, for example 202601MM-202607MM. Quarterly ranges use SS, for example 202601SS-202602SS. Use time codes supported by the selected dataset. Statistics queries accept at most 10 indicator IDs and 5 time ranges. Region is fixed to nationwide (000000000000).

Output / 输出

Field	Meaning
action	Executed operation
name	Upstream indicator or catalog display name
cid	Dataset ID, when available
indicatorId	Indicator UUID; not a catalog node ID
nodeId	Catalog node ID, for catalog results
period, periodName	Time code and display label
value	Original upstream value; strings and nulls are preserved
unit	Upstream unit if supplied as a separate field; otherwise null (search titles may contain units)
region	Region name, when supplied
sourceUrl, retrievedAt	Source homepage and retrieval timestamp
raw	Original metadata for this record

A statistics dataset item represents one indicator in one reporting period. Search, catalog, and indicator-list items each represent one returned row. We do not reinterpret cumulative values as quarterly values, infer units from names, or replace missing values with zero. The source link identifies the official website, not a deep link to a specific table.

One run saves at most maxResults records (default 100, maximum 1000). OUTPUT in the default key-value store summarizes the upstream row count, selected count, stored count, truncation, and dataset ID. It does not expose additional unsaved records. Results may be fewer if a platform spending limit is reached. An empty query saves no dataset items.

Pricing / 计费口径

Consult this Actor's current Pricing tab for actual charges. If per-result pricing is enabled, one result means one saved dataset row, as defined above, not one whole query or one dataset. A query returning 2 periods for 3 indicators may create 6 results. Any enabled Actor-start charge is separate and can apply even when a query fails or is empty. No custom charging event is emitted by this code; the default dataset-item event is handled by Apify.

计费以平台 Pricing 页面为准。若启用按结果计费，统计查询按保存的“指标 × 报告期”记录数计费；搜索、目录、指标列表按实际保存行数计费。空结果和错误不会创建结果记录，但平台设置的启动费用仍可能产生。

Limits and errors / 限制与错误

• Nationwide data only; no provincial/city comparison, forecasting, or seasonal adjustment performed by this Actor.
• No NBS API key or Cloudflare Worker credentials required. Requests go directly to the official data endpoint.
• Upstream requests time out after 20 seconds and responses are limited to 5 MiB. Narrow the time range or indicator selection if needed.
• Invalid input, HTTP failures, non-JSON challenge pages, upstream business errors, and unexpected data structures fail the run. Details are written to the ERROR key-value record and run status.
• Availability, catalog IDs, published values, and upstream response formats can change. A successful search on another platform does not guarantee that Apify's network can reach the source.
• No proxy or browser challenge bypass is included. If access is restricted, report the error through this Actor's Issues tab.

API and MCP

Use Apify Console's generated API examples to run this Actor with the JSON inputs above. The Actor can also be invoked through Apify's MCP integration. It is a task that runs and exits; it does not host a persistent /mcp endpoint or use Standby mode. The repository's Cloudflare Worker separately retains its Streamable HTTP /mcp service.