Technical

In Taiwan stock technical data, we have 20 datasets, as follows:

Taiwan Stock Overview TaiwanStockInfo¶

This table mainly lists all Taiwan listed (TWSE), OTC (TPEx), and Emerging stocks, including stock names, codes, and industry categories!
Data update time: 1:30 daily. The actual update time is based on the API data.

Transferred stocks keep multiple rows: take the row with the latest date as the current market

If a stock moved from the Emerging board to TWSE / TPEx, this table keeps both an emerging row and a twse / tpex row. The date of the emerging row is frozen at the day the stock left the Emerging board, while the twse / tpex row carries the current date. Filtering with type == "emerging" alone therefore also picks up stocks that have already transferred; take the row with the latest date for each stock_id instead. This matters especially for TaiwanStockPriceTick and TaiwanStockKBar, whose volume unit differs by market (lots for TWSE / TPEx, shares for Emerging).

Example

PackagePython-requestR

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_info()

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockInfo",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset = "TaiwanStockInfo"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	industry_category	stock_id	stock_name	type	date
0	ETF	0050	元大台灣50	twse	2021-10-05
1	ETF	0051	元大中型100	twse	2021-10-05
2	ETF	0052	富邦科技	twse	2021-10-05
3	ETF	0053	元大電子	twse	2021-10-05
4	ETF	0054	元大台商50	twse	2021-10-05

{
    industry_category: str, # industry category
    stock_id: str, # stock code
    stock_name: str, # stock name
    type: str, # market type
    date: str # update date
}

Taiwan Stock Overview (with Warrants) TaiwanStockInfoWithWarrant¶

This table mainly lists all Taiwan listed (TWSE), OTC (TPEx), and Emerging stocks and warrants, including names, codes, and industry categories!
Total record count exceeds 50,000.
Data update time: 1:30 daily. The actual update time is based on the API data.

Example

PackagePython-requestR

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_info_with_warrant()

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockInfoWithWarrant",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset = "TaiwanStockInfoWithWarrant"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	industry_category	stock_id	stock_name	type	date
0	ETF	0050	元大台灣50	twse	2021-10-05
1	ETF	0051	元大中型100	twse	2021-10-05
2	ETF	0052	富邦科技	twse	2021-10-05
3	ETF	0053	元大電子	twse	2021-10-05
4	ETF	0054	元大台商50	twse	2021-10-05

{
    industry_category: str, # industry category
    stock_id: str, # stock code
    stock_name: str, # stock name
    type: str, # market type
    date: str # update date
}

Provides the warrant underlying reference table. (Since it includes warrant data, the data volume is large and takes about a minute.)
Covers both listed (TWSE) and OTC (TPEX) warrants' underlying (target_stock_id) reference; OTC warrant underlying history goes back to 2011-01-03, so you can look up the warrants of a given underlying (including expired and code-reused historical warrants).
Data update time: 1:30 daily. The actual update time is based on the API data.

A single warrant code may map to different underlying stocks

Warrants have an expiration date. Once a warrant code (stock_id) expires and is delisted, that code is recycled and reissued to a new warrant on a (possibly different) underlying stock. As a result, the same stock_id may appear in multiple rows in this table, each mapping to a different underlying (target_stock_id) and each with its own date (listing date) and end_date (last trading date).

To determine which underlying a warrant code maps to on a given day, match by "the query date falls within that row's date ~ end_date interval"; reused codes from different periods are naturally distinguished by their non-overlapping intervals.

Example

Python-requestR

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockInfoWithWarrantSummary",
    "data_id": "2330",# optional parameter
    "start_date": "2020-04-06",# optional parameter
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset = "TaiwanStockInfoWithWarrantSummary",
        data_id = "2330",# optional parameter
        start_date = "2020-04-06"# optional parameter
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	stock_id	date	target_stock_id	type	fulfillment_method	end_date	fulfillment_start_date	fulfillment_end_date	exercise_ratio
0	052018	2023-07-18	2330	認購	美式	2025-07-15	2023-07-18	2025-07-17	0.1
1	052405	2023-07-21	2330	認購	美式	2025-07-17	2023-07-21	2025-07-21	0.01
2	057397	2023-09-04	2330	認購	美式	2025-09-01	2023-09-04	2025-09-03	0.03
3	057607	2023-09-06	2330	認購	美式	2025-09-03	2023-09-06	2025-09-05	0.06
4	060985	2023-10-03	2330	認購	美式	2025-09-30	2023-10-03	2025-10-02	0.03

{
    stock_id: str, # stock code
    date: str, # listing date
    close: float, # closing price
    target_stock_id: str, # underlying stock code
    target_close: float, # underlying closing price
    type: str, # warrant type
    fulfillment_method: str, # exercise method
    end_date: str, # last trading date
    fulfillment_start_date: str, # exercise start date
    fulfillment_end_date: str, # exercise end date
    exercise_ratio: float, # exercise ratio
    fulfillment_price: float # exercise price
}

Taiwan Stock Trading Date TaiwanStockTradingDate¶

Provides Taiwan stock trading dates.
Data update time: Monday to Friday 18:00. The actual update time is based on the API data.

Example

Python-requestR

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockTradingDate",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset = "TaiwanStockTradingDate"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	date
0	2005-01-03
1	2005-01-04
2	2005-01-05
3	2005-01-06
4	2005-01-07

{
    date: str # trading date
}

Daily Stock Price Information TaiwanStockPrice¶

Provides daily trading information for Taiwan listed (TWSE), OTC (TPEx), and Emerging stocks!
Data range: 1994-10-01 ~ now
Data update time: Monday to Friday 17:30. The actual update time is based on the API data.

Emerging stocks: open is the previous-day average price, and may fall outside the day's high/low (not a data error)

Emerging-board stocks trade by negotiation and have no opening price. For emerging stocks this table therefore fills open with the previous-day average price, not an opening price; max / min / close are still the day's high / low / last traded price and are correct.

As a result, an emerging stock's open can sit above max or below min on volatile days. This is a field-definition difference, not corrupt data. For candlestick charts, use max / min / close, or treat open as the previous-day average.

Example: 6515 穎崴 on 2020-03-19 (emerging at the time) → open=218.06, max=200, min=170, close=174.9; 218.06 is exactly the previous-day average that day. The stock later moved to TWSE, after which its data is normal.

A related case: open=0 on an emerging stock's first trading day. Since an emerging stock's open is the previous-day average, on a stock's very first trading day there is no previous day to average, so open shows as 0; max / min / close are still the day's correct values, and from the next day onward open carries a normal previous-day average. This too is a field-definition difference, not corrupt data.

Example: 8272 全景 on 2023-10-17 (its first day on the emerging board) → open=0, max=136, min=112.5, close=130; the next day, 2023-10-18, open=129.39, which is 2023-10-17's average price. The same rule holds for TaiwanStockPriceAdj (adjusted price): the adjusted price is the raw price times an adjustment factor, so 0 carries through unchanged.

When no traded price is published for a stock on a given day, open / max / min / close are all 0 (not a data error)

In the daily quote published by TWSE / TPEx, a stock with no published traded price for that day has its open/high/low/close shown as --. This table follows the original publication and converts that to the numeric value 0, so open / max / min / close / spread are all 0 for that stock on that day. The source publication itself carries no price — this is not a crawling gap.

This can happen in any market (TWSE, TPEx, and Emerging alike) and is unrelated to the emerging-board open definition described above. A halted stock, a day with no trades at all, or a day with only sporadic trades that produce no published price all fall into this category.

Example 1 (no trades at all): 2317 鴻海 on 2025-07-30 → Trading_Volume=0, and open/max/min/close are all 0; the TWSE publication for that day literally reads "volume 0, open/high/low/close --". The previous trading day 2025-07-29 (close=171.5) and the next one 2025-07-31 (close=178) are both normal.

Example 2 (volume present but no published price): 9929 秋雨 on 2025-07-31 → Trading_Volume=451, yet open/max/min/close are still 0; the TWSE publication reads "volume 451, open/high/low/close --".

How to detect it: because of Example 2, do not filter on Trading_Volume > 0 alone — that condition misses rows that have volume but no price. Filter on the price columns instead:

df = df[df["close"] > 0]          # keep only days with a published traded price
# or forward-fill from the previous trading day's close
df["close"] = df["close"].replace(0, method="ffill")

Difference from TaiwanStockPriceAdj: the adjusted-price table is not 0 on these days — it carries forward the previous trading day's adjusted price (while Trading_Volume keeps the day's actual value). For instance, 2317 鴻海 on 2025-07-30 in TaiwanStockPriceAdj is open=169.94, max=170.43, min=166.06, close=166.54, i.e. 2025-07-29's adjusted price. Use TaiwanStockPriceAdj if you want a series with no 0 gaps; use TaiwanStockPrice's 0 if you need to identify which days had no published price.

Example

PackagePython-requestRPackage-Async

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_daily(
    stock_id='2330',
    start_date='2020-04-02',
    end_date='2020-04-12'
)

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockPrice",
    "data_id": "2330",
    "start_date": "2020-04-02",
    "end_date": "2020-04-12",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockPrice",
        data_id= "2330",
        start_date= "2020-04-02",
        end_date= "2020-04-08"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

from FinMind.data import DataLoader
from loguru import logger
import datetime

api = DataLoader()
api.login_by_token(api_token='token')

start = datetime.datetime.now()
df = api.taiwan_stock_daily(
    stock_id_list=['2330', '2317', '2454', '3008'],
    start_date='2024-01-01',
    end_date='2024-12-31',
    use_async=True,
)
cost = datetime.datetime.now() - start
logger.info(cost)

Output

DataFrameSchema

	date	stock_id	Trading_Volume	Trading_money	open	max	min	close	spread	Trading_turnover
0	2020-04-06	2330	59712754	16324198154	273	275.5	270	275.5	4	19971
1	2020-04-07	2330	48887346	13817936851	283.5	284	280.5	283	7.5	24281
2	2020-04-08	2330	38698826	11016972354	285	285.5	283	285	2	19126
3	2020-04-09	2330	29276430	8346209654	287.5	288	282.5	283	-2	15271
4	2020-04-10	2330	28206858	7894277586	280	282	279	279.5	-3.5	15833

{
    date: str, # date
    stock_id: str, # stock code
    Trading_Volume: int64, # trading volume
    Trading_money: int64, # trading amount
    open: float64, # open price
    max: float64, # max price
    min: float64, # min price
    close: float64, # close price
    spread: float64, # price change
    Trading_turnover: float32 # number of trades
}

Example

PackagePython-requestR

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_daily(
    start_date='2020-04-06',
)

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockPrice",
    "start_date": "2020-04-06",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockPrice",
        start_date= "2020-04-06"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	date	stock_id	Trading_Volume	Trading_money	open	max	min	close	spread	Trading_turnover
0	2020-04-06	0050	12207626	935731083	76.95	77.1	75.75	77.05	1.15	5824
1	2020-04-06	0051	33000	953030	29.05	29.05	28.74	29.05	0.38	21
2	2020-04-06	0052	178700	10660088	59.4	60.05	58.75	60	1.25	56
3	2020-04-06	0053	17000	589750	34.66	35	34.48	34.84	0.18	17
4	2020-04-06	0054	10000	200040	19.87	20.03	19.87	20.03	0	4

{
    date: str, # date
    stock_id: str, # stock code
    Trading_Volume: int64, # trading volume
    Trading_money: int64, # trading amount
    open: float64, # open price
    max: float64, # max price
    min: float64, # min price
    close: float64, # close price
    spread: float64, # price change
    Trading_turnover: int64 # number of trades
}

Provides daily trading information for Taiwan listed (TWSE), OTC (TPEx), and Emerging stocks!
Data range: 2000-01-01 ~ now
Data update time: Monday to Friday 17:30. The actual update time is based on the API data.

Example

PackagePython-requestRPackage-Async

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_weekly(
    stock_id='2330',
    start_date='2020-04-02',
    end_date='2020-04-12'
)

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockWeekPrice",
    "data_id": "2330",
    "start_date": "2020-04-02",
    "end_date": "2020-04-12",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockWeekPrice",
        data_id= "2330",
        start_date= "2020-04-02",
        end_date= "2020-04-08"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

from FinMind.data import DataLoader
from loguru import logger
import datetime

api = DataLoader()
api.login_by_token(api_token='token')

start = datetime.datetime.now()
df = api.taiwan_stock_weekly(
    stock_id_list=['2330', '2317', '2454', '3008'],
    start_date='2024-01-01',
    end_date='2024-12-31',
    use_async=True,
)
cost = datetime.datetime.now() - start
logger.info(cost)

Output

DataFrameSchema

	stock_id	yweek	max	min	trading_volume	trading_money	trading_turnover	date	close	open	spread
0	2330	2020W15	288	270	409564428	114799189198	188964	2020-04-06	279.5	273	8

{
    stock_id: str, # stock code
    yweek: str, # weekly date
    max: float64, # max price
    min: float64, # min price
    trading_volume: int64, # trading volume
    trading_money: int64, # trading amount
    trading_turnover: float32, # number of trades
    date: str, # date
    close: float64, # close price
    open: float64, # open price
    spread: float64 # price change
}

Example

Python-requestR

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockWeekPrice",
    "start_date": "2020-04-06",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockWeekPrice",
        start_date= "2020-04-06"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	stock_id	yweek	max	min	trading_volume	trading_money	trading_turnover	date	close	open	spread
0	2330	2020W15	288	270	409564428	114799189198	188964	2020-04-06	279.5	273	8

{
    stock_id: str, # stock code
    yweek: str, # weekly date
    max: float64, # max price
    min: float64, # min price
    trading_volume: int64, # trading volume
    trading_money: int64, # trading amount
    trading_turnover: float32, # number of trades
    date: str, # date
    close: float64, # close price
    open: float64, # open price
    spread: float64 # price change
}

Provides daily trading information for Taiwan listed (TWSE), OTC (TPEx), and Emerging stocks!
Data range: 2000-01-01 ~ now
Data update time: Monday to Friday 17:30. The actual update time is based on the API data.

Example

PackagePython-requestRPackage-Async

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_monthly(
    stock_id='2330',
    start_date='2020-04-02',
    end_date='2020-04-12'
)

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockMonthPrice",
    "data_id": "2330",
    "start_date": "2020-04-02",
    "end_date": "2020-04-12",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockMonthPrice",
        data_id= "2330",
        start_date= "2020-04-02",
        end_date= "2020-05-08"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

from FinMind.data import DataLoader
from loguru import logger
import datetime

api = DataLoader()
api.login_by_token(api_token='token')

start = datetime.datetime.now()
df = api.taiwan_stock_monthly(
    stock_id_list=['2330', '2317', '2454', '3008'],
    start_date='2024-01-01',
    end_date='2024-12-31',
    use_async=True,
)
cost = datetime.datetime.now() - start
logger.info(cost)

Output

DataFrameSchema

	stock_id	ymonth	max	min	trading_volume	trading_money	trading_turnover	date	close	open	spread
0	2330	2020M05	301.5	288.5	1744651784	513799591970	788158	2020-05-01	292	294.5	-12.5

{
    stock_id: str, # stock code
    ymonth: str, # monthly date
    max: float64, # max price
    min: float64, # min price
    trading_volume: int64, # trading volume
    trading_money: int64, # trading amount
    trading_turnover: float32, # number of trades
    date: str, # date
    close: float64, # close price
    open: float64, # open price
    spread: float64 # price change
}

Example

Python-requestR

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockMonthPrice",
    "start_date": "2020-04-06",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockMonthPrice",
        start_date= "2020-04-06"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	stock_id	ymonth	max	min	trading_volume	trading_money	trading_turnover	date	close	open	spread
0	2330	2020M05	301.5	288.5	1744651784	513799591970	788158	2020-05-01	292	294.5	-12.5

{
    stock_id: str, # stock code
    ymonth: str, # monthly date
    max: float64, # max price
    min: float64, # min price
    trading_volume: int64, # trading volume
    trading_money: int64, # trading amount
    trading_turnover: float32, # number of trades
    date: str, # date
    close: float64, # close price
    open: float64, # open price
    spread: float64 # price change
}

Data range: 1994-10-01 ~ now
Data update time: Monday to Friday 17:30. The actual update time is based on the API data.

When the adjustment factor updates, and which day is the baseline

Adjusted prices are computed backwards: the most recent trading day in the data is the baseline, and adjustment factors are applied cumulatively to earlier days. As a result, the adjusted price on the event day itself always equals that day's raw price — the adjustment shows up in the history before that day.

For ex-dividend / ex-rights, capital reduction, stock split and par value change events, the adjustment factor is applied in the post-close update on the event day itself; it is not deferred to the next trading day.

If a scheduled event date falls on a market-wide closure (a typhoon suspension, for example) and is postponed, the adjustment follows the day trading actually took place, not the originally scheduled date.

Example

PackagePython-requestRPackage-Async

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_daily_adj(
    stock_id='2330',
    start_date='2020-04-02',
    end_date='2020-04-12'
)

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockPriceAdj",
    "data_id": "2330",
    "start_date": "2020-04-02",
    "end_date": "2020-04-12",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockPriceAdj",
        data_id= "2330",
        start_date= "2020-04-02",
        end_date= "2020-04-08"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

from FinMind.data import DataLoader
from loguru import logger
import datetime

api = DataLoader()
api.login_by_token(api_token='token')

start = datetime.datetime.now()
df = api.taiwan_stock_daily_adj(
    stock_id_list=['2330', '2317', '2454', '3008'],
    start_date='2024-01-01',
    end_date='2024-12-31',
    use_async=True,
)
cost = datetime.datetime.now() - start
logger.info(cost)

Output

DataFrameSchema

	date	stock_id	Trading_Volume	Trading_money	open	max	min	close	spread	Trading_turnover
0	2020-04-06	2330	59712754	16324198154	273	275.5	270	275.5	4	19971
1	2020-04-07	2330	48887346	13817936851	283.5	284	280.5	283	7.5	24281
2	2020-04-08	2330	38698826	11016972354	285	285.5	283	285	2	19126
3	2020-04-09	2330	29276430	8346209654	287.5	288	282.5	283	-2	15271
4	2020-04-10	2330	28206858	7894277586	280	282	279	279.5	-3.5	15833

{
    date: str, # date
    stock_id: str, # stock code
    Trading_Volume: int64, # trading volume
    Trading_money: int64, # trading amount
    open: float64, # open price
    max: float64, # max price
    min: float64, # min price
    close: float64, # close price
    spread: float64, # price change
    Trading_turnover: float32 # number of trades
}

Example

Python-requestR

import requests
import pandas as pd
url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockPriceAdj",
    "start_date": "2020-04-06",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockPriceAdj",
        start_date= "2020-04-06"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = data$data %>%
do.call('rbind',.) %>%
data.table
head(df)

Output

DataFrameSchema

	date	stock_id	Trading_Volume	Trading_money	open	max	min	close	spread	Trading_turnover
0	2020-04-06	0050	12207626	935731083	76.95	77.1	75.75	77.05	1.15	5824
1	2020-04-06	0051	33000	953030	29.05	29.05	28.74	29.05	0.38	21
2	2020-04-06	0052	178700	10660088	59.4	60.05	58.75	60	1.25	56
3	2020-04-06	0053	17000	589750	34.66	35	34.48	34.84	0.18	17
4	2020-04-06	0054	10000	200040	19.87	20.03	19.87	20.03	0	4

{
    date: str, # date
    stock_id: str, # stock code
    Trading_Volume: int64, # trading volume
    Trading_money: int64, # trading amount
    open: float64, # open price
    max: float64, # max price
    min: float64, # min price
    close: float64, # close price
    spread: float64, # price change
    Trading_turnover: float32 # number of trades
}

(Due to the large data volume, each request only provides one stock's data for one day.)

Data range: 2018-12-07 ~ now.
Providing the dataset, stock_id, and start_date parameters returns data for that day.
Data update time: Monday to Friday 15:30. The actual update time is based on the API data.
Some data is missing on these dates: 2018-12-22, 2019-02-20, 2019-02-21, 2019-02-22. In addition, 2019-05-16 only has data for a small number of ETFs.
Enabling Async significantly reduces the data update time. In a Colab test, downloading 2,236 stocks took only 3 minutes 40 seconds.

TickType has been corrected across the full history (2018-12-07 ~ 2023-03-10)

Before 2023-03-13, trades whose side could not be determined were all labelled 1 (buyer-initiated), which made the buyer-initiated share noticeably too high. Every trading day in this range has been regenerated using the same rule applied from 2023-03-13 onward. If you downloaded any of these dates before, please fetch the data again.

For dates before 2021-06-22, about 8% ~ 10% of trades have TickType 0 (undetermined), because the source data itself carries no side information for those trades, while the old version labelled them all as buyer-initiated. If you group statistics by TickType, treat 0 separately rather than merging it into either side.

The unit of volume differs by market: lots for TWSE / TPEx, shares for Emerging

The volume column keeps each market's native trading unit: TWSE and TPEx stocks are in lots (1 lot = 1,000 shares); Emerging stocks are in shares. Both units therefore appear in the same day of data. This is a difference between markets, not a data error. The same rule applies to volume in TaiwanStockKBar (minute K).

To tell which market a stock currently belongs to, use the type column of TaiwanStockInfo (twse / tpex / emerging). Note: a stock that moved from Emerging to TWSE / TPEx keeps both rows in TaiwanStockInfo — an emerging row and a twse / tpex row — and the date of the emerging row is frozen at the day the stock left the Emerging board. Filtering with type == "emerging" alone therefore also picks up stocks that have already moved, which makes the Emerging volume unit look inconsistent. Take the row with the latest date for each stock_id as the current market.

Example: for 1294, the emerging row has date 2024-09-25 while the tpex row carries the current date, so it is a TPEx stock today and its volume is in lots; 1260 has only an emerging row carrying the current date, so it is still an Emerging stock and its volume is in shares.

Also note, when reconciling tick data against the daily volume in TaiwanStockPrice: for TWSE / TPEx stocks the daily volume includes block trades (see TaiwanStockBlockTrade), while tick data does not, so the two will not match exactly.

Example

PackagePython-requestPackage-AsyncR

from FinMind.data import DataLoader

api = DataLoader()
# api.login_by_token(api_token='token')
df = api.taiwan_stock_tick(
    stock_id='2330',
    date='2020-01-02'
)

import io
import requests
import pandas as pd

url = "https://api.finmindtrade.com/api/v4/data"
token = "" # Refer to login to obtain the token
headers = {"Authorization": f"Bearer {token}"}
parameter = {
    "dataset": "TaiwanStockPriceTick",
    "data_id": "2330",
    "start_date": "2020-01-02",
}
resp = requests.get(url, headers=headers, params=parameter)
data = resp.json()
data = pd.DataFrame(data["data"])
print(data.head())

from FinMind.data import DataLoader
from loguru import logger
import datetime

token = ""

data_loader = DataLoader()
data_loader.login_by_token(token)

date = '2025-12-08'
start = datetime.datetime.now()
df = data_loader.taiwan_stock_tick(
    date=date,
    use_async=True,
)
cost = datetime.datetime.now() - start
logger.info(cost)
# 00:03:41

library(httr)
library(data.table)
library(dplyr)
url = 'https://api.finmindtrade.com/api/v4/data'
token = "" # Refer to login to obtain the token
response = httr::GET(
    url = url,
    query = list(
        dataset="TaiwanStockPriceTick",
        data_id= "2330",
        start_date= "2020-01-02"
    ),
    add_headers(Authorization = paste("Bearer", token))
)
data = content(response)
df = do.call('cbind',data$data) %>%
data.table
head(df)

Output

DataFrameSchema

	date	stock_id	deal_price	volume	Time
0	2020-01-02	2330	332.5	520	09:00:00.000
1	2020-01-02	2330	332.5	520	09:00:00.646
2	2020-01-02	2330	333	45	09:00:05.000
3	2020-01-02	2330	333	45	09:00:05.660
4	2020-01-02	2330	333	22	09:00:10.000

{
    date: str, # date
    stock_id: str, # stock code
    deal_price: float64, # deal price
    volume: int64, # volume
    Time: str, # time
    TickType: str # tick type (0: undetermined, 1: buyer-initiated (trade at ask), 2: seller-initiated (trade at bid))
}

Fetch all data for a specific date at once (available only to sponsorpro members)¶

(Due to the large data volume, each request only provides one day's data.)

Data range: 2018-12-07 ~ now.
Providing the dataset and date parameters returns data for that day.
Data update time: Monday to Friday 15:30. The actual update time is based on the API data.
Some data is missing on these dates: 2018-12-22, 2019-02-20, 2019-02-21, 2019-02-22. In addition, 2019-05-16 only has data for a small number of ETFs.