Page 12 - MONECO Financial Training Catalogue
P. 12
PYTHON FOR INVESTMENT PROFESSIONALS
• Visualize portfolio diversification structure When to Use Bootstrap:
• Identify clusters of correlated assets • Small sample sizes
• Network-based risk analysis • Non-normal distributions
• Understand which stocks move together • Metrics without closed-form formulas
• Regulatory applications (systemic risk) • Communicating uncertainty to non-technical audiences
Case Study: Visual Portfolio Diversification Analysis WEDNESDAY, MAY 20
with JPMorgan Long-Term Capital Market Assumptions, 00 30
Stability of Diversification Over Time 09 –12
• Calculate correlation matrices Performance Attribution & Analysis
• Transform correlations to distances
• Build minimum spanning trees Module 13: Return Attribution
• Visualize portfolio structure Case Study: Sector-Based Performance Attribution
• Identify diversification opportunities Background: Decompose active return into allocation
• Detect asset clusters and selection effects
• Calculate diversification metrics • Perform Brinson-Fachler attribution
• Decompose active returns
• Interpret MST for portfolio construction
• Analyze sector contributions
Risk Measurement & Bootstrap Analysis • Create attribution reports
• Identify sources of alpha
Module 11: VaR & CVaR – Risk Metrics
Case Study: Portfolio Risk Dashboard with Historical, Module 14: Performance Analytics Suite
Parametric and Monte Carlo VaR, CvaR, VaR Limit Case Study: Automated Performance Report
Breaches, Backtesting VaR Models (Kupiec Test) • Build reusable metric functions
• Calculate VaR using three methods • Calculate comprehensive performance stats
• Compute CVaR/Expected Shortfall • Automate monthly reporting
• Understand when each method is appropriate • Create tearsheets
• Backtest VaR models
• Visualize risk metrics Professional Reporting & Visualization
Module 15: Creating Fund Factsheets
Module 12: Bootstrap Methods & Confidence Intervals
Why Bootstrapping? Case Study: Automated Monthly Factsheet
• Understand uncertainty in risk metrics • Create multi-panel factsheets
• Generate professional charts
• Don’t assume normality (returns aren’t normal!)
• Generate confidence bands for any metric • Export publication-quality images
• Regulatory stress testing applications • Build repeatable templates
• Client communication: “here’s the range of possible • HTML, PDF output
outcomes”
Module 16: Excel Report Generation
Case Study A: Direct Data from Stock Exchanges Case Study: Formatted Excel Reports
• Generate multi-sheet workbooks
Case Study B: Bootstrap Confidence Intervals for
Maximum Drawdown • Apply Excel formatting from Python
• Create professional reports
The Problem: • Automate monthly processes
• Max drawdown is a single historical number
• But what’s the uncertainty around it? 12 –13
30
30
• Could it have been worse with different ordering? Lunch break
30
• Bootstrap answers: “What’s the range of likely 13 –17 30
drawdowns?” Production-Ready Scripts & Automation
Case Study C: Bootstrap for Multiple Risk Metrics
Module 17: From Notebook to Script
Practical Applications: Why Scripts vs Notebooks:
• Risk disclosure to investors • Notebooks = Exploration
• Regulatory stress testing • Scripts = Production
• Capital allocation decisions • Scripts can be scheduled and automated
• Performance fee hurdle rates
• Client reporting with uncertainty Case Study: Production Portfolio Analytics Script
• Structure code with functions
• Add error handling
12
