Skip to content

README.md

Latest commit

 

History

History
124 lines (99 loc) · 6.09 KB

README.md

File metadata and controls

124 lines (99 loc) · 6.09 KB

Macroeconomic Indicators & Market Volatility Analysis

Investment Strategy Development Through Quantitative Risk Modeling

Executive Summary

Bottom Line Up Front: Current market environment supports cautious bullish positioning with 75% equity allocation, reduced from typical 85% due to overbought technical conditions (RSI: 71.2). Predicted monthly volatility: 2.5% with 95% VaR at -4.2%.

Business Impact Delivered:

  • Portfolio managers receive tactical allocation signals with quantified risk parameters
  • Risk teams gain enhanced volatility predictions (1.5% accuracy vs 3-4% industry standard)
  • Client advisory teams access clear risk-return scenarios with historical validation
  • Investment committee receives regime-based allocation framework reducing drawdown risk

Key Findings

Metric S&P 500 NASDAQ Insight
Annual Return 7.2% 9.0% NASDAQ premium comes with 43% higher volatility
Sharpe Ratio 0.340 0.324 S&P 500 delivers superior risk-adjusted returns
Max Drawdown -52.6% -75.0% Diversification reduces tail risk by 22 percentage points
95% VaR -8.0% -10.1% S&P 500 monthly loss threshold 2.1% lower

Market Regime Analysis: Bull/Low Volatility conditions occur 51% of time and generate optimal risk-adjusted returns. Current regime classification supports equity overweight with tactical caution.

Investment Recommendations

Portfolio Allocation

  • S&P 500: 55% (core broad market exposure)
  • NASDAQ: 20% (growth component with volatility awareness)
  • Bonds: 15% (defensive allocation for rate environment)
  • Cash: 10% (tactical reserve for opportunities)

Risk Management Framework

  • Monthly VaR Estimate: -4.2% (95% confidence level)
  • Rebalancing Triggers: 5% allocation drift or regime classification changes
  • Position Sizing: Low volatility environment supports normal position sizes
  • Downside Protection: Stop-loss protocols activated if predicted volatility exceeds 6%

Rationale

Allocation reflects overbought technical conditions requiring defensive positioning despite supportive Bull/Low Vol regime. 10% cash reserve enables opportunistic rebalancing when RSI normalizes below 70.

Technical Implementation

Data Architecture

Sources: Federal Reserve Economic Data (FRED), Yahoo Finance API
Coverage: 307 monthly observations (2000-2025)
Indicators: Unemployment rate, inflation, federal funds rate, Treasury yields, VIX, market indices
Quality Controls: Timezone standardization, missing value treatment, outlier detection

Analytical Framework

Phase 1: Data Foundation

  • Multi-source data integration (FRED, Yahoo Finance)
  • Timezone standardization and frequency alignment
  • Quality assurance and missing value treatment
  • Data validation and outlier detection

Phase 2: Financial Analysis

  • Correlation matrices and statistical relationships
  • Risk metrics (VaR, Sharpe ratios, drawdowns)
  • Market regime classification (Bull/Bear × High/Low Vol)
  • Technical indicator calculations

Phase 3: Predictive Modeling

  • Random Forest volatility prediction models
  • Feature engineering with lagged variables
  • Cross-validation and performance testing
  • Model interpretation and feature importance

Phase 4: Investment Strategy

  • Portfolio optimization and allocation recommendations
  • Scenario analysis and stress testing
  • Executive reporting and visualization
  • Risk management framework development

Model Performance

  • Volatility Prediction: Random Forest achieving 1.54% Mean Absolute Error
  • Top Predictors: 3-month moving average (30.9%), lagged returns (7.4%), VIX indicators (16.1%)
  • Regime Classification: Bull/Bear identification with volatility overlay
  • Backtesting: Strategy framework tested across multiple market cycles

Historical Strategy Performance

Total Return Comparison: Buy & Hold (363.3%) vs Tactical Strategy (210.1%)
Key Insight: Simple buy-and-hold outperformed tactical timing by 153 percentage points, demonstrating market timing difficulty and validating disciplined allocation approach over reactive strategies.

Repository Structure

  • 01_Data_Collection_Validation.ipynb - Multi-source data integration and quality assurance
  • 02_EDA_Financial_Metrics.ipynb - Risk metrics and correlation analysis
  • 03_Time_Series_Risk_Modeling.ipynb - Predictive modeling and technical indicators
  • 04_Executive_Dashboard.ipynb - Investment strategy and reporting
  • README.md - Project documentation
  • requirements.txt - Python environment setup

Technologies Used

Core: Python, pandas, numpy, scikit-learn
Financial Data: yfinance, pandas-datareader (FRED API)
Analysis: scipy, statsmodels for statistical modeling
Visualization: matplotlib, seaborn for professional charts
Machine Learning: Random Forest regression for volatility forecasting

Methodology Highlights

  • Risk Metrics: Industry-standard VaR, Expected Shortfall, Maximum Drawdown calculations
  • Correlation Analysis: Pearson correlation matrices with statistical significance testing
  • Technical Analysis: RSI, moving averages, Bollinger Bands for market timing
  • Regime Classification: Bull/Bear markets with volatility overlays for tactical allocation
  • Predictive Modeling: Supervised learning with lagged features and cross-validation

Limitations & Assumptions

  • Analysis focused on US markets; international diversification not modeled
  • Risk-free rate assumed at 2% for Sharpe ratio calculations
  • Transaction costs simplified in backtesting scenarios
  • Historical relationships may not persist in changing market structures

Data Access

Raw datasets are not included in this repository due to size constraints. The notebooks automatically download data from:

  • Federal Reserve Economic Data (FRED) API
  • Yahoo Finance API
  • Original Kaggle dataset reference provided in notebook 01

Contact