Fundamentals — Core Financial Statements
As-reported core financial statements for A-share listed companies — balance sheet, income statement and cash flow, exactly as each filing disclosed them, with a strict point-in-time guarantee. Two tables: the filing dimension and the as-reported fact panel.
Buy by the field
2 priced fields
Pick the fields you need — priced individually below.
Exactly the sum of its fields — no bundle premium, no bundle discount
Pay annually · save 20% — $2,800/yr
ⓘ Take the whole set from the field list below — select everything, add it to your basket, and check out there.
Data due diligence
- Point-in-time: yes
- Carries the disclosure date; restatements preserved.
- Personal data: none
- No personal information.
- Web collection: no
- No web collection; sourced from filings or a licensed feed.
- Generative AI: no
- No third-party generative AI is used; processing is proprietary ML/NLP.
The personal-data values are descriptive, not risk grades: no personal information in any dataset is collected from individuals; all of it is drawn from mandatory public company disclosure or already-published news.
Declarations published by ChinaScope. Open the Due diligence tab and sign in to read the full questionnaire.
Fundamentals — Core Financial Statements
fin_statementsSource DataAs-reported core financial statements for A-share listed companies — balance sheet, income statement and cash flow, exactly as each filing disclosed them, with a strict point-in-time guarantee. Two tables: the filing dimension and the as-reported fact panel.
Daily source stitch · data frontier 2026-10-28
11. Field reference: every column
The full field dictionary: every table, every column, with its type and meaning. One real sample is shown per table (Kweichow Moutai, 600519, Q3 2025 balance sheet).
filing
Table · pick one of 2
filing
16 columns · grain (primary key): report_id
| Column | Name | Type | Null | Description |
|---|---|---|---|---|
operation | CDC operation flag | string | yes | A record-maintenance flag marking whether this row was added (A), updated (U), or deleted (D) in the latest delivery, so you can apply incremental changes correctly when keeping your copy in sync. |
report_id ·PK | Filing / disclosure ID | string | no | ChinaScope's unique identifier for a single filed statement: one company, one fiscal period, one schedule. It is the primary key of this table and the join key that every line item in as_reported points back to. |
secu | Security code | string | yes | ChinaScope's standardized stock code for the filer's listed security; the suffix marks the listing venue (_SH_EQ Shanghai, _SZ_EQ Shenzhen, _KC_EQ STAR Market, _BJ_EQ Beijing). Resolves to the security master for ~99.6% of filings; a small pre-listing/provisional tail is _TEMP/_KC. |
company_id | Company ID (CSF master) | string | no | ChinaScope's company code; resolves 100% to the company master (base_company). This is the company key: aggregate and join on it, not on secu. |
publish_date | Public-record (disclosure) date | date | no | The date the statement was filed and became public: the point-in-time anchor. Filter publish_date <= D for what was public on date D; to also exclude ChinaScope's later corrections, use _known_date as well (§8). |
report_date | Report end date | date | no | The fiscal period-end (closing) date the statement covers, e.g. 2025-09-30 for Q3 2025. |
report_period | Reporting period (month) | string | yes | The period-end month: 3 (Q1), 6 (interim/H1), 9 (Q3), or 12 (annual). |
report_quarter | Reporting quarter | string | yes | The reporting milestone: Q1, Q2 (interim), Q3, or Q4 (annual). |
fiscal_year | Fiscal year-end (MM-DD) | string | yes | The month-and-day the company's fiscal year closes; 12-31 for A-shares. |
currency | Reporting currency | string | yes | The currency in which the statement's figures are denominated (CNY for A-shares). |
rpt_code | Statement schedule code | string | yes | Which schedule this filing is: 2.1.1 income statement, 2.2.1 balance sheet, 2.3.1 cash flow (main), 2.4.1 cash-flow supplementary (indirect reconciliation), 2.5.1 non-recurring items, 2.6.1 per-share / ROE (§6). |
is_valid | Valid flag | boolean | yes | A validity flag; a filing must be valid (and carry a non-look-ahead disclosure date) for its line items to be admitted at read time. |
create_time | Record create timestamp | timestamp | yes | When this record was first created in the dataset. |
update_time | Record update timestamp | timestamp | yes | When this record was most recently updated. |
_known_date | Delivery date | date | no | The date of the ChinaScope delivery this version of the row arrived in; the panel is partitioned by it. It is not the disclosure date (that is publish_date on filing): it is the second date a correction-aware view needs, used to keep each row's latest version delivered on or before your date (§8). Every row of the history carries 2026-04-13, a marker one day before the history's delivery on 2026-04-14. |
_panel_op | Applied change marker | string | no | How this version of the row was applied to the panel: A added, U updated, D deleted. It can differ from the delivered operation flag: a row delivered as an update whose key had never been seen is an insert and is marked A here. To exclude deleted rows, first keep each key's latest version (§8), then drop it if that version is 'D'; dropping 'D' rows alone leaves the earlier versions of a deleted row in place. |
Sample row:
operation: A
report_id: 600519_SH_EQCSF00000015982.2.12025-09-30…
secu: 600519_SH_EQ
company_id: CSF0000001598
publish_date: 2025-10-30
report_date: 2025-09-30
report_period: 9
report_quarter: Q3
fiscal_year: 12-31
currency: CNY
rpt_code: 2.2.1 (balance sheet)
is_valid: True
Sourced as-delivered from ChinaScope, with resolved codes and structural links computed by Numinor. Licensing passes through to you; fields are served exactly as the vendor issues them.
Access & FAQ
- How do I buy just a few fields?
- Check the fields you want and you'll see your running subtotal. Select every field and the whole-set price applies automatically. Add them to your basket and check out — this dataset, or fields combined across several.
- Can I evaluate these fields in the Matrix before I buy?
- Yes — load the actual source tables in the Matrix and analyze them with an AI agent before you buy. See the Overview tab for this dataset’s Matrix evaluation lag; “None” means no intentional delay.
- How is it delivered?
- Apache Parquet over a signed-URL REST API, the same pipe as Construct Data. The free Matrix sandbox serves an evaluation snapshot. Its delay, if any, is shown in the Overview tab.
- When is each update available, and how reliable is it?
- Source data refreshes on the ChinaScope cadence and is delivered through the same signed-URL pipeline as Construct Data, with the same manifest/status behaviour. Evaluate freshness yourself in the Matrix before you commit.
- Why are some fields free and others cost thousands?
- Price tracks non-replicability. Resolved codes and structural links are the moat; machine scores and raw text are cheap; identifier keys are free. Figures are priced per table; codes are charged once per dataset.
Still have questions? Contact Sales →
Questions This Dataset Comes Up In
Public reference pages answering a question that spans several datasets, this one among them. No account needed.
- Why a join to China A-share company data silently drops rowsEntity resolution in Chinese company data is partial, and the resolved share differs by dataset and by relationship type. An inner join discards exactly the rows that were not resolved, and those rows are not a random sample.
- Which China A-share datasets are point-in-time, and which only look like itA dataset is point-in-time only if it tells you when a figure became public, not just which period it covers. Several widely used China datasets carry the period and not the filing date, which reads as point-in-time and is not.
- Missing, zero, or blank: three different states in China filing dataA real zero, a blank cell the filer left empty, and a line with no row at all are three distinct states in filing data. Collapsing them to one fills a panel with zeros that were never reported.
- Code and unit traps in China A-share dataSentiment where 2 means negative, ownership ratios in two different units inside one dataset, and standardized line items that hide the as-reported label. Each one returns plausible numbers when read wrongly.
- One company, many securities: why A-share company counts come out too highCompany facts attach to each listed security, so a company with two listings appears twice. Grouping on the ticker instead of the company key inflates every company-level aggregate.