luminpulse-ai/quant-analysis

Quantitative Analysis Skill

Description

Perform quantitative finance research including data analysis, portfolio optimization, risk modeling, and econometric analysis.

Tools Used

• jupyter_execute - Execute Python code for financial analysis (auto-switches to Jupyter)
• jupyter_notebook - Manage analysis notebooks
• update_notebook - Set up analysis cells in Jupyter
• update_latex - Write finance paper content to LaTeX editor
• latex_compile - Compile research papers (auto-switches to LaTeX editor)
• update_notes - Write analysis summaries and findings

Capabilities

Data Analysis

• Time series analysis of financial returns
• Cross-sectional regression (Fama-MacBeth, panel data)
• Event studies and abnormal return analysis
• Volatility modeling (GARCH family)

Portfolio Optimization

• Mean-variance optimization (Markowitz)
• Black-Litterman model with views
• Risk parity and equal risk contribution
• Factor-based portfolio construction

Risk Analysis

• Value-at-Risk (VaR) and Conditional VaR
• Stress testing and scenario analysis
• Copula-based dependency modeling
• Monte Carlo simulation

Usage Patterns

Analyze Returns

When user says: "Analyze the performance of [asset/portfolio]"

1. Load price data using pandas/yfinance
2. Calculate returns, volatility, Sharpe ratio
3. Plot cumulative returns and drawdowns
4. Run statistical tests (normality, autocorrelation)
5. Present findings with charts

Build a Model

When user says: "Build a [pricing/risk/factor] model"

1. Clarify model specification and data requirements
2. Load and clean data
3. Estimate model parameters
4. Validate with out-of-sample testing
5. Report results with diagnostics

Tool Examples

Load and analyze stock returns

# via jupyter_execute
import yfinance as yf
import pandas as pd
import numpy as np

data = yf.download("AAPL", start="2023-01-01", end="2024-01-01")
returns = data["Close"].pct_change().dropna()
print(f"Mean: {returns.mean():.4f}, Vol: {returns.std():.4f}, Sharpe: {returns.mean()/returns.std()*np.sqrt(252):.2f}")

Validation checkpoints

• Verify data has no missing values or extreme outliers before modeling
• Check model residuals for autocorrelation after estimation
• Confirm out-of-sample period has no look-ahead bias