stock-correlation

---

name: stock-correlation description: > Analyze stock correlations to find related companies and trading pairs. Use when the user asks about correlated stocks, related companies, sector peers, trading pairs, or how two or more stocks move together. Triggers: "what correlates with NVDA", "find stocks related to AMD", "correlation between AAPL and MSFT", "what moves with", "sector peers", "pair trading", "correlated stocks", "when NVDA drops what else drops", "stocks that move together", "beta to", "relative performance", "supply chain partners", "correlation matrix", "co-movement", "related tickers", "sympathy plays", "semiconductor peers", "hedging pair", "realized correlation", "rolling correlation", or any request about stocks that move in tandem or inversely. Also triggers for well-known pairs like AMD/NVDA, GOOGL/AVGO, LITE/COHR. If only one ticker is provided, infer the user wants correlated peers.

Stock Correlation Analysis Skill

Finds and analyzes correlated stocks using historical price data from Yahoo Finance via yfinance. Routes to specialized sub-skills based on user intent.

Important: This is for research and educational purposes only. Not financial advice. yfinance is not affiliated with Yahoo, Inc.

---

Step 1: Ensure Dependencies Are Available

Current environment status:

!`python3 -c "import yfinance, pandas, numpy; print(f'yfinance={yfinance.__version__} pandas={pandas.__version__} numpy={numpy.__version__}')" 2>/dev/null || echo "DEPS_MISSING"`

If DEPS_MISSING, install required packages before running any code:

import subprocess, sys
subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "yfinance", "pandas", "numpy"])

If all dependencies are already installed, skip the install step and proceed directly.

---

Step 2: Route to the Correct Sub-Skill

Classify the user's request and jump to the matching sub-skill section below.

User Request	Route To	Examples
Single ticker, wants to find related stocks	Sub-Skill A: Co-movement Discovery	"what correlates with NVDA", "find stocks related to AMD", "sympathy plays for TSLA"
Two or more specific tickers, wants relationship details	Sub-Skill B: Return Correlation	"correlation between AMD and NVDA", "how do LITE and COHR move together", "compare AAPL vs MSFT"
Group of tickers, wants structure/grouping	Sub-Skill C: Sector Clustering	"correlation matrix for FAANG", "cluster these semiconductor stocks", "sector peers for AMD"
Wants time-varying or conditional correlation	Sub-Skill D: Realized Correlation	"rolling correlation AMD NVDA", "when NVDA drops what else drops", "how has correlation changed"

If ambiguous, default to Sub-Skill A (Co-movement Discovery) for single tickers, or Sub-Skill B (Return Correlation) for two tickers.

Defaults for all sub-skills

Parameter	Default
Lookback period	1y (1 year)
Data interval	1d (daily)
Correlation method	Pearson
Minimum correlation threshold	0.60
Number of results	Top 10
Return type	Daily log returns
Rolling window	60 trading days

---

Sub-Skill A: Co-movement Discovery

Goal: Given a single ticker, find stocks that move with it.

A1: Build the peer universe

You need 15-30 candidates. Do not use hardcoded ticker lists — build the universe dynamically at runtime. See references/sector_universes.md for the full implementation. The approach:

1. Screen same-industry stocks using yf.screen() + yf.EquityQuery to find stocks in the same industry as the target
2. Broaden to sector if the industry screen returns fewer than 10 peers
3. Add thematic/adjacent industries — read the target's longBusinessSummary and screen 1-2 related industries (e.g., a semiconductor company → also screen semiconductor equipment)
4. Combine, deduplicate, remove target ticker

A2: Compute correlations

import yfinance as yf
import pandas as pd
import numpy as np

def discover_comovement(target_ticker, peer_tickers, period="1y"):
    all_tickers = [target_ticker] + [t for t in peer_tickers if t != target_ticker]
    data = yf.download(all_tickers, period=period, auto_adjust=True, progress=False)

    # Extract close prices — yf.download returns MultiIndex (Price, Ticker) columns
    closes = data["Close"].dropna(axis=1, thresh=max(60, len(data) // 2))

    # Log returns
    returns = np.log(closes / closes.shift(1)).dropna()
    corr_series = returns.corr()[target_ticker].drop(target_ticker, errors="ignore")

    # Rank by absolute correlation
    ranked = corr_series.abs().sort_values(ascending=False)

    result = pd.DataFrame({
        "Ticker": ranked.index,
        "Correlation": [round(corr_series[t], 4) for t in ranked.index],
    })
    return result, returns

A3: Present results

Show a rank