load_data.py

import yfinance as yf
import pandas as pd
import os

FILENAME = "market_data.parquet"  # Parquet is much faster and smaller than CSV


def get_data(tickers, start_date, end_date):
    # Check if we already have the file
    if os.path.exists(FILENAME):
        print("Loading data from local disk...")
        return pd.read_parquet(FILENAME)

    # Otherwise, download it
    print("Downloading data from yfinance...")
    df = yf.download(tickers, start=start_date, end=end_date, auto_adjust=True)

    if not df.empty:
        # Save it for next time
        df.to_parquet(FILENAME)
        return df
    else:
        raise ValueError("Download failed and no local file found.")


# Usage
tickers = ["AAPL", "MSFT", "GOOGL", "AMZN", "NVDA", "META", "TSLA"]
sector_map = {}
for ticker in tickers:
    try:
        sector_map[ticker] = yf.Ticker(ticker).info.get('sector', 'Unknown')
    except Exception:
        sector_map[ticker] = 'Unknown'
df = get_data(tickers, "2013-01-01", "2024-12-31")
df = df.stack(level=1).rename_axis(['Date', 'Ticker']).sort_index()  # cleaner for returns
g = df.groupby('Ticker')