Taiwan Warrants Daily Quotes — 權證行情 API avatar

Taiwan Warrants Daily Quotes — 權證行情 API

Pricing

from $0.50 / 1,000 results

Go to Apify Store
Taiwan Warrants Daily Quotes — 權證行情 API

Taiwan Warrants Daily Quotes — 權證行情 API

Every listed and OTC warrant in Taiwan, priced daily: ~40,000 warrants a session with OHLC, volume, the underlying's own close, and the strike, exercise ratio and expiry joined on — plus moneyness, intrinsic value, premium and leverage computed from them. TWSE sessions reach back to 2004-02-11.

Pricing

from $0.50 / 1,000 results

Rating

0.0

(0)

Developer

chris

chris

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

3 days ago

Last modified

Share

Taiwan Warrants Daily Quotes (權證行情) API

Get structured JSON access to every warrant listed in Taiwan, priced daily — around 31,000 on the TWSE and 9,500 on the TPEX in a single session, each with its OHLC, volume and turnover, the underlying's own closing price alongside it, and the strike, exercise ratio and expiry joined on from the exchange's issue register. TWSE sessions go back to 2004-02-11.

Taiwan runs one of the largest listed warrant markets in the world by instrument count. On 2026-08-18 there were 40,665 live warrants across both boards — more than thirty for every listed company — written by fourteen brokers over 1,300 underlyings. 14,593 of them actually traded that day. The exchanges publish this as six separate category reports plus a 40 MB issue register, all in Chinese, with no way to ask for one underlying's chain. This Actor assembles it into one record shape, cross-checks it against feeds the exchange compiles independently, and lets you ask the question everyone actually has: what warrants are there on 2330, and what are they worth?

Why this data matters

  • underlyingCodes turns a 31,000-row session into an option chain. Ask for ["2330"] and you get the 462 TSMC warrants that traded on 2026-08-18, each with its strike, expiry and moneyness — the view no exchange page offers.
  • The analytics are computed, not scraped, so they are internally consistent. moneyness_pct, intrinsic_value, time_value, premium_pct and leverage_ratio all come from the same three published numbers: underlying close, exercise price and exercise ratio. Nothing is modelled or fitted, and a missing input yields null rather than an assumption.
  • The exercise ratio is the trap this dataset defuses. TWSE publishes it as shares per thousand warrant units and TPEX as shares per unit. Get it wrong and 3,209 of the 14,593 warrants that traded on 2026-08-18 appear to close below their own intrinsic value. Both are normalised to per-unit here, and the check that proves it runs on every session.
  • trades_below_intrinsic finds the genuine anomalies. After correct scaling, exactly 2 warrants closed below intrinsic value on 2026-08-18 — both at or one day from their last trading day. That is a real, and rare, market artefact rather than a parsing error, so it is flagged instead of smoothed away.
  • Warrants expire, and this dataset says exactly when. days_to_expiry and last_trading_date come from the issue register, not from the code. The name's own expiry hint (5A = October 2025) is kept in expiry_code for reference and is not what the dates are computed from.
  • issue_notes is an audit trail nobody else publishes. For TWSE warrants it is the issuer's own running log of every strike adjustment and every cancellation: "◎2026/07/09標的證券除息,調整後履約價格1,475.41元". It explains why exercise_price differs from original_exercise_price.
  • Nobody serves this in English. Search for a Taiwan warrant API and you find nothing free and nothing structured. The source is six Chinese category reports behind a host that resets the connection if you ask too quickly.

Output

One record per warrant per trading session. A TSMC call, deep in the money with two days left:

{
"date": "2026-08-18",
"market": "TWSE",
"warrant_code": "032840",
"warrant_name": "台積電中信58購02",
"warrant_type": "CALL",
"warrant_type_zh": "認購權證",
"issuer": "中信",
"expiry_code": "58",
"suspended": false,
"knock_out_price": null,
"open": 7.25,
"high": 7.25,
"low": 7.25,
"close": 7.25,
"price_change": null,
"volume_units": 5000,
"transactions": 1,
"turnover": 36250,
"best_bid": 7.15,
"best_bid_size": 20,
"best_ask": null,
"best_ask_size": 0,
"underlying_code": "2330",
"underlying_name": "台積電",
"underlying_close": 2380,
"exercise_price": 1480.27,
"original_exercise_price": 1500,
"exercise_ratio": 0.008,
"expiry_date": "2026-08-20",
"last_trading_date": "2026-08-18",
"exercise_start_date": "2026-08-20",
"days_to_expiry": 2,
"exercise_style": null,
"category": "一般型",
"cap_price": null,
"floor_price": null,
"issued_units": 10000000,
"market_maker_quoting": "主動報價",
"settlement_style_code": "2",
"terms_as_of": "2026-08-18",
"terms_matched": true,
"moneyness_pct": 60.7815,
"intrinsic_value": 7.19784,
"time_value": 0.05216,
"premium_pct": 0.2739,
"leverage_ratio": 2.6262,
"trades_below_intrinsic": false
}

price_change is null here because the exchange marked the session X — not comparable with the previous close, which happens after a strike adjustment or a resumption. An unknown change is reported as unknown rather than as zero.

A far out-of-the-money put on the same underlying, where the leverage is the whole point:

{
"warrant_code": "03002T",
"warrant_name": "台積電群益5A售12",
"warrant_type": "PUT",
"issuer": "群益",
"close": 0.34,
"underlying_close": 2380,
"exercise_price": 1855.05,
"exercise_ratio": 0.019,
"days_to_expiry": 59,
"moneyness_pct": -28.2984,
"intrinsic_value": 0.0,
"time_value": 0.34,
"premium_pct": 22.8086,
"leverage_ratio": 133.0
}

Set includeTerms: false for a quotes-only run: everything from date through underlying_close, without the 40 MB issue-register download.

Input

FieldTypeNotes
marketsarrayTWSE, TPEX. Default both.
startDate / endDatestringYYYY-MM-DD. Both empty means the latest session. TWSE reaches back to 2004-02-11; 90 calendar days per run.
warrantTypesarrayCALL, PUT, BULL, BEAR, EXTENDABLE_BULL, EXTENDABLE_BEAR. Each is a separate request, so narrowing saves time.
underlyingCodesarraye.g. ["2330", "2317"]. The filter most callers want.
warrantCodesarraye.g. ["03812T"]. Matched as text — codes are not all numeric.
issuersarrayThe abbreviation in the warrant's name: 元大, 凱基, 群益, 統一, 永豐, 國票, 富邦, 台新, 國泰, 中信, 元富, 兆豐, 第一, 康和.
onlyTradedbooleanDrop zero-volume warrants. Roughly halves the output.
includeTermsbooleanJoin strike/expiry and compute analytics. Default true.
crossCheckbooleanVerify against independently compiled feeds; results in the log. Default true.

How the numbers are verified

Every run checks itself against sources the exchange builds separately, and says so in the log:

  • Turnover against t187ap42_L. A second TWSE feed publishing the same session. All 31,116 warrants agree on volume and turnover, exactly, on the day this was written. Its column is labelled 成交張數 ("lots") but holds units — the agreement is what proves it.
  • Underlying closes against STOCK_DAY_ALL. The equity closing feed, compiled independently of the warrant reports. All 30,179 share-linked warrants match to the cent. The remaining 937 are written on indices (IX0001 臺股指數 and friends), which no equity feed carries.
  • TPEX against its own mirror. tpex_warrant_quts republishes tpex_warrant_daily_quts under a second endpoint. All 9,549 rows agree.
  • OHLC and turnover consistency, per row. low ≤ open, close ≤ high, and turnover is non-zero exactly when volume is. Every row, every session.
  • Column signatures, per report. The six TWSE reports use two different twenty-column layouts — calls and puts open with 暫停交易, while bull/bear certificates drop it and carry 牛熊證觸及限制價格 instead. Both are pinned; anything else fails the run loudly rather than silently mis-assigning columns.

Coverage and limits

  • TWSE history reaches 2004-02-11 and no further — the exchange answers 查詢日期小於93年2月11日 before that. The column layout has been stable across every session tested from 2004 to today.
  • TPEX publishes only the latest session. There is no dated TPEX warrant endpoint; the OTC feeds are today-only snapshots. Ask for a historical range and the run logs that TPEX was skipped rather than quietly returning nothing.
  • Issue terms are a live snapshot, for both boards. terms_as_of records when. For a historical session the strike shown is today's — after every dividend adjustment since — and warrants that have expired since will have terms_matched: false because they have left the register. For the latest session the match rate is 31,116 / 31,116.
  • issuer is read off the warrant's name, which the exchange builds to a fixed rule. Fourteen abbreviations cover every warrant on both boards; anything else is left null rather than guessed.
  • leverage_ratio is plain financial gearing — spot × ratio / warrant price. It is not delta-adjusted effective gearing, which would require a pricing model this Actor deliberately does not run.
  • A full TWSE session takes several minutes. The call report alone is 4.8 MB served uncompressed, and requests are paced to avoid the connection resets TWSE throws at fast callers.

Taiwan Market Data Suite

This Actor is part of a suite of 23 Taiwan market data APIs by chamarix — official sources only, cross-validated against independent official endpoints, clean JSON out. Code samples for the whole suite: GitHub.

Market data:

Property market:

Government & civic data:

Data source

Taiwan Stock Exchange 每日收盤行情 (MI_INDEX, warrant categories) and 權證發行相關資料 (t187ap37_L); Taipei Exchange 上櫃權證收盤行情日報表 and 上櫃權證發行基本資料. Public data, no login required.