Where can we get fundamental data for NSE stocks? Data sources mentioned mostly cater for US market

General
1

Course Name: [Getting Market Data: Stocks, Crypto, News

2

Free API:

	import yfinance as yf
	# Define the stock ticker
	ticker = "RELIANCE.NS"  # NSE-listed stock
	stock = yf.Ticker(ticker)
	# Get basic info
	info = stock.info
	print("Company Info:", info)
	# Get financials
	financials = stock.financials
	print("Financials:", financials)
	# Get balance sheet
	balance_sheet = stock.balance_sheet
	print("Balance Sheet:", balance_sheet)
	Paid API:
	Capital Market

Others:

However, note that QuantInsti is not affiliated with these vendors, and before getting the data from any vendor, it is essential that traders carry out necessary research about the vendor.

3

Thanks

4

@Ajay_Pawar yfinance is giving empty dataframe as output

5

Can you update yfinance and try again with:


import yfinance as yf

ticker = "RELIANCE.NS"  # Replace with your stock symbol
stock = yf.Ticker(ticker)

# Income Statement
income_stmt = stock.financials
print(income_stmt)
6

Thanks it is working

7

@Ajay_Pawar Can you please let me know how can LLMs be used to extract information from the filings? Are there any resources that can guide me through this?

8

Hi Shreyas,

We do have blog on implementing the LLM : https://blog.quantinsti.com/application-llm-portfolio-management-thematic-index/
You can try:

import openai

openai.api_key = "your-api-key"

file = openai.files.create(
    file=open("example_10k.pdf", "rb"),
    purpose="assistants"
)

assistant = openai.beta.assistants.create(
    name="Filing Extractor",
    instructions="Extract specific financial data from uploaded company filings and return it as a Python dictionary.",
    tools=[{"type": "retrieval"}],
    model="gpt-4-1106-preview"
)

thread = openai.beta.threads.create()

openai.beta.threads.messages.create(
    thread_id=thread.id,
    role="user",
    content="""
From the attached filing, extract the following fields and return them in a Python dictionary:

- Company Name
- Filing Date
- Fiscal Year
- Revenue
- Net Income
- Risk Factors (as a list)
- Summary of Management Discussion
""",
    file_ids=[file.id]
)

run = openai.beta.threads.runs.create(
    thread_id=thread.id,
    assistant_id=assistant.id
)

However you will have to develop method to validate the extracted numbers.
like sampling, compare with benchmark etc.

9

Okay thanks.