Holdings changes
/v1/fund-holding-changesInstitutionalCompares the most recent quarterly reports and returns lists of new entries, dropouts, adds and cuts, plus a multi-period trend. Institutional and above.
Answers: What did this fund add, cut, or drop this quarter?
Request parameters
| Parameter | Required | Type | Default | Description |
|---|---|---|---|---|
code | ✅ | string | — | 6-digit open-end fund code, e.g. 110022 |
periods | — | int | 4 | How many reporting periods to look back (2–8) |
top_n | — | int | 10 | Compare the top N holdings (5–20) |
Response fields
| Field | Type | Description |
|---|---|---|
latest_period / prev_period | string | Latest / previous reporting period |
top_n_effective | int | The comparison range actually used (bounded by disclosures in each period) |
summary | object | Totals: new_in / out / increase / decrease / flat / top_weight_pct / disclosed / disclosed_prev |
changes[] | array | Per-stock changes, see below |
changes[].action | string | Change type, e.g. new_in / out / increase / decrease / flat |
changes[].code / name | string | Stock code / name |
changes[].weight_pct / weight_prev_pct / weight_delta_pct | number | Current / previous share of NAV and the delta |
changes[].shares_wan / shares_prev_wan / shares_delta_pct | number | Shares held (10k shares) and the change |
changes[].market_value_wan | number | Market value held this period (10k CNY, as of period end) |
history[] | array | Multi-period trend: period / disclosed / top_weight_pct / new_in / out / increase / decrease |
Real response sample
Captured live using 110022. List fields are shortened for readability (the real response is longer); elapsed_ms is from that one call — the first request usually takes 1–3 seconds.
{
"ok": true,
"endpoint": "fund-holding-changes",
"tier": "institutional",
"elapsed_ms": 21752,
"data": {
"code": "110022",
"name": "易方达消费行业股票",
"scale_yi": 101.09,
"latest_period": "2026年3季度",
"prev_period": "2026年2季度",
"top_n": 10,
"top_n_effective": 10,
"summary": {
"new_in": 0,
"out": 0,
"increase": 3,
"decrease": 0,
"flat": 7,
"replaced_pct": 0,
"top_weight_pct": 72.45,
"top_weight_prev_pct": 67.6,
"top_weight_delta_pct": 4.85,
"disclosed": 13,
"disclosed_prev": 48
},
"changes": [
{
"action": "increase",
"code": "002594",
"name": "比亚迪",
"weight_pct": 8.64,
"weight_prev_pct": 7.96,
"weight_delta_pct": 0.68,
"shares_wan": 1048.87,
"shares_prev_wan": 969.5,
"shares_delta_pct": 8.19,
"market_value_wan": 87381.67
},
{
"action": "increase",
"code": "000596",
"name": "古井贡酒",
"weight_pct": 6.84,
"weight_prev_pct": 4.45,
"weight_delta_pct": 2.39,
"shares_wan": 738.95,
"shares_prev_wan": 541.1,
"shares_delta_pct": 36.56,
"market_value_wan": 69129.03
},
{
"action": "increase",
"code": "000568",
"name": "泸州老窖",
"weight_pct": 5.35,
"weight_prev_pct": 5,
"weight_delta_pct": 0.35,
"shares_wan": 747.58,
"shares_prev_wan": 626.45,
"shares_delta_pct": 19.34,
"market_value_wan": 54034.9
}
],
"history": [
{
"period": "2026年3季度",
"disclosed": 13,
"top_weight_pct": 72.45,
"new_in": 0,
"out": 0,
"increase": 3,
"decrease": 0
},
{
"period": "2026年2季度",
"disclosed": 48,
"top_weight_pct": 67.6,
"new_in": 2,
"out": 2,
"increase": 3,
"decrease": 3
},
{
"period": "2026年1季度",
"disclosed": 13,
"top_weight_pct": 67.24,
"new_in": 2,
"out": 2,
"increase": 1,
"decrease": 6
}
],
"method": "对比定义:只在两期都完整披露的前 10 名范围内逐只比对,股数变化是数据源披露的持股数(万股)直接相减,占净值比变化是数据源披露的占净值比例直接相减;replaced_pct = (新进数 + 掉出数) / (2 × 10)。这些都是本接口基于披露持仓算出来的,不是数据源给的现成字段。",
"note": "季报(1/3 季度)只披露前十大,半年报/年报披露全部持仓,所以本期 disclosed 会比半年报小很多,这是披露规则不是清仓。另外,送股、转增、拆分会让股数被动变化,本接口不做除权还原;基金调仓也可能在同一天完成买回,前十大快照看不出来。"
}
}Example call
curl "https://api.fundapi.cn/v1/fund-holding-changes?code=110022" \ -H "Authorization: Bearer YOUR_KEY" \ -H "User-Agent: your-app"
Notes
Disclosure counts vary widely — always read `disclosed`
Quarterly reports disclose only the top 10; semi-annual reports disclose everything. The sample fund disclosed 13 holdings this period and 48 last period. A naive set difference would read that as "sold 35 stocks".
`out` does not mean liquidated
It only means "outside the disclosed range this period" — not that the fund sold everything. To judge a real exit, check whether shares held fell to zero.