chore: initial project scaffold for quant trading learning

Sets up project structure with yfinance-based OHLCV fetcher for top 100 S&P companies, Jupyter notebook scaffold, and uv-managed deps. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-25 00:58:02 +08:00
commit 60fa0c113d
7 changed files with 5406 additions and 0 deletions
@@ -0,0 +1,67 @@
+import pandas as pd
+import yfinance as yf
+import pyarrow.parquet as pq
+from pathlib import Path
+import time
+
+
+TOP100_TICKERS = [
+    "AAPL", "MSFT", "NVDA", "AMZN", "META", "GOOGL", "GOOG", "TSLA", "BRK-B", "JPM",
+    "LLY", "V", "UNH", "XOM", "MA", "AVGO", "PG", "HD", "COST", "JNJ",
+    "ABBV", "MRK", "PEP", "KO", "CVX", "NFLX", "ADBE", "CRM", "AMD", "ACN",
+    "MCD", "WMT", "BAC", "CSCO", "TMO", "ABT", "DHR", "CMCSA", "UPS", "LIN",
+    "NEE", "PM", "TXN", "HON", "ORCL", "QCOM", "RTX", "INTU", "AMGN", "LOW",
+    "AMAT", "BMY", "UNP", "AXP", "CAT", "BKNG", "SPGI", "VRTX", "DE", "GILD",
+    "ELV", "ISRG", "ADP", "MDT", "LMT", "CVS", "CI", "ADI", "SYK", "SBUX",
+    "MMC", "C", "TJX", "GS", "BLK", "SCHW", "ZTS", "ETN", "PLD", "BDX",
+    "REGN", "ADI", "CL", "CME", "CB", "SO", "DUK", "NOC", "FIS", "ITW",
+    "APD", "NSC", "BSX", "GD", "FI", "MMM", "PGR", "WM", "AFL", "SLB",
+]
+
+
+def fetch_top100_ohlcv(output_path: str, period: str = "max") -> None:
+    output_path = Path(output_path)
+    output_path.mkdir(parents=True, exist_ok=True)
+
+    print(f"Fetching OHLCV data for top 100 companies (period={period})...")
+    records = []
+    errors = []
+
+    for i, ticker in enumerate(TOP100_TICKERS, 1):
+        try:
+            print(f"[{i}/{len(TOP100_TICKERS)}] Downloading {ticker}...", end=" ")
+            hist = yf.Ticker(ticker).history(period=period, auto_adjust=False)
+            if hist.empty:
+                print(f"no data, skipping")
+                errors.append((ticker, "no data"))
+                continue
+            hist = hist.reset_index()
+            hist["Date"] = hist["Date"].astype(str)
+            hist["ticker"] = ticker
+            records.append(hist)
+            print(f"{len(hist)} rows")
+            time.sleep(0.1)
+        except Exception as e:
+            print(f"error: {e}")
+            errors.append((ticker, str(e)))
+
+    if not records:
+        print("No data fetched.")
+        return
+
+    df = pd.concat(records, ignore_index=True)
+    cols = ["ticker", "Date", "Open", "High", "Low", "Close", "Volume", "Dividends", "Stock Splits"]
+    cols = [c for c in cols if c in df.columns]
+    df = df[cols]
+
+    parquet_path = output_path / "top100_companies.parquet"
+    df.to_parquet(parquet_path, index=False)
+    print(f"\nSaved {len(df)} rows to {parquet_path}")
+    print(f"Date range: {df['Date'].min()} -> {df['Date'].max()}")
+    print(f"Tickers fetched: {df['ticker'].nunique()}")
+    if errors:
+        print(f"Errors ({len(errors)}): {errors[:10]}")
+
+
+if __name__ == "__main__":
+    fetch_top100_ohlcv("data/top100_ohlcv")