Building an AI agent? Start with /llms.txt for the full site index.

Workflows / Use Cases

Researching company fundamentals

Building a verifiable Taiwan fundamentals workflow from monthly revenue, the income statement, the balance sheet and valuation data.

Who this is for

Developers, researchers and agent integrators building a Taiwan fundamentals workflow on the API.

If what you want is a research process that reruns and can be checked, start here.

Which datasets fundamentals research uses

Fundamentals work usually draws on monthly revenue, the income statement, the balance sheet, the cash flow statement and valuation data. Use one consistent rule for tickers, dates and sourcing across all of them — cross-dataset misalignment is the most common way this goes wrong.

Prerequisites

Before starting:

Python 3.9+
requests
X-API-Key
A symbol, for example 2330

Authentication

Every request carries X-API-Key. The examples below read different datasets through the same helper function.

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {
    "X-API-Key": "your_api_key_here",
}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

Querying the issuer profile

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {"X-API-Key": "your_api_key_here"}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

profile = get_dataset(
    "/v2/datasets/issuer-profile",
    {"symbol": "2330"},
)

print(profile)

Querying monthly revenue

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {"X-API-Key": "your_api_key_here"}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

monthly_revenue = get_dataset(
    "/v2/datasets/monthly-revenue",
    {
        "symbol": "2330",
        "limit": 12,
    },
)

print(monthly_revenue)

Querying the income statement

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {"X-API-Key": "your_api_key_here"}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

income_statement = get_dataset(
    "/v2/datasets/income-statement",
    {"symbol": "2330", "limit": 4},
)

print(income_statement)

Querying the balance sheet

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {"X-API-Key": "your_api_key_here"}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

balance_sheet = get_dataset(
    "/v2/datasets/balance-sheet",
    {"symbol": "2330", "limit": 4},
)

print(balance_sheet)

Querying the cash flow statement

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {"X-API-Key": "your_api_key_here"}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

cash_flow = get_dataset(
    "/v2/datasets/cash-flow-statement",
    {"symbol": "2330", "limit": 4},
)

print(cash_flow)

Querying valuation data

import requests

BASE_URL = "https://api.twmarketdata.com"
HEADERS = {"X-API-Key": "your_api_key_here"}

def get_dataset(path, params):
    response = requests.get(
        f"{BASE_URL}{path}",
        headers=HEADERS,
        params=params,
    )
    response.raise_for_status()
    return response.json()

valuation_data = get_dataset(
    "/v2/datasets/valuation-data",
    {"symbol": "2330", "limit": 4},
)

print(valuation_data)

Assembling a fundamentals summary

In practice: read the change in monthly revenue first, then the income statement and balance sheet, and use valuation data last to relate price back to the fundamentals.

company_name = (profile.get("rows") or [{}])[0].get("company_name")
latest_revenue_yoy = (monthly_revenue.get("rows") or [{}])[0].get("yoy_growth_pct")
latest_eps = (income_statement.get("rows") or [{}])[0].get("eps")
latest_per = (valuation_data.get("rows") or [{}])[0].get("per")
latest_pbr = (valuation_data.get("rows") or [{}])[0].get("pbr")

summary = {
    "company_name": company_name,
    "latest_revenue_yoy": latest_revenue_yoy,
    "latest_eps": latest_eps,
    "latest_per": latest_per,
    "latest_pbr": latest_pbr,
}

print(summary)

Data gaps and freshness

Datasets update on different cadences and disclose at different times. Before using them together, check each response's time fields, its range and its data_gaps.

TW Market Data preserves source role and gap information. Do not assume any dataset covers every date for every ticker.

Next steps

With a fundamentals summary in place, the natural extensions are: