Asset Allocation and Portfolio Construction

This programme covers the latest trends in quantitative modelling for asset allocation and portfolio construction and highlights approaches which help improving real-world in investment decision processes by taking into account risk factors, capital market scenarios, estimation risk and generally real-world aspects in applied investing beyond mean-variance.

Innovations suggested over the last twenty years are contrasted with current industry practice and illustrated with examples with an eye for practical implementation. Perspectives are provided on the latest industry trends like Smart Beta and Machine Learning / Artificial Intelligence.

Six practical exercises simulate real-life key decisions in asset allocation and portfolio construction. Mathematical concepts are discussed and illustrated using Excel spreadsheets that delegates can take away.

• Quants / Financial engineers
• Chief Investment Officers
• Portfolio and investment managers
• Fund and wealth managers
• Treasury and liquidity managers
• Risk managers
• Traders
• Strategists

• Comprehensive discussion of different approaches to portfolio construction
• Latest research trends and industry best practice and implications for practitioners
• Practical suggestions on estimation risk, model risk and advanced topics
• Workshops on all models and detailed teaching on quantitative concepts

• Basic knowledge of financial markets, asset classes and derivative instruments
• Elementary mathematics and statistics (probability distributions mean, variance and correlation)
• Microsoft Excel

Review of MPT and Traditional Asset Allocation Practice

• The impact of the Financial Crisis and the low-yield environment on investment management
• Review of Modern Portfolio Theory (MPT) - discussion of pre-course readings and questions
  • The many definitions of “risk”
  • A reinterpretation of Two-Fund-Separation as factor investing advice
  • MPT applications: surplus management, active management relative to a benchmark
• Issues of traditional asset allocation industry practice
  • Pseudo asset classes & diversification
  • Risk characteristics of fixed weight policy portfolios
  • Drivers of the success of rebalancing strategies
  • Industry Trend: Factor-based asset allocation

Exercise: Factor risk budgeting in a mean-variance framework

An Overview of Newer Approaches to Asset Allocation

• Investment horizons and time-variable risk and return characteristics
• The case for flexible asset allocations: moving from static portfolio risk to managing portfolio risk and return dynamically over time
• Industry Trend “Smart Beta”: Is it a strategy, a product or a distribution approach?

Expected Returns & Return-Based Asset Allocation

• Return-Based Strategies:
  • Market-cap-weighted passive & adaptive asset allocation
  • Tactical asset allocation, strategic asset allocation
• Estimating expected returns
  • Scenario-based approaches to return estimation: Markov-Regime switching models and approaches used in the industry
  • Alpha estimation: benchmark-implied returns, Treynor/Black portfolios, Information Coefficient
  • The Black/Litterman model & introduction to Bayesian Expectations
  • Denoising with the James Stein return estimator
• Turning forecast scores into portfolio allocations without optimizers
  • Turning rankings into scores
  • Deriving risk-adjusted scores
  • Taking into account restrictions: tactical bands, turnover, general linear constraints

Exercise: A macroeconomic scenario-based approach to risk and return estimation

Risk Expectations and Risk-Based Investing

• Risk-based investment strategies
  • Minimum variance
  • Risk parity
  • Risk budgeting
  • Equal-weighted
  • Most-diversified, maximum diversification
  • Drivers of the success of low risk investment strategies: interest rate regimes, market anomalies or factor risk premium?
• Time-varying risk characteristics
  • Autocorrelation & volatility clustering
  • Comparison of volatilities and correlation dynamics
  • Understanding the positive relationship between risk and return
• Volatility and correlation matrix estimation
  • Historical estimators, rolling and exponentially-weighted implementations, introduction to GARCH
  • Properties of a valid correlation matrix, consequences of invalid correlation matrices
  • Robust correlation estimators: average correlation matric, Bayesian shrinkage estimators (the Ledoit/Wolf approach)
  • Handling asymmetrical dependencies and tail-risk dependence

Exercise: Developing a quantitative volatility trading strategy

Estimation Risk and Its Management

• A scenario-approach to explicitly taking into account uncertainty in expected returns and correlations
• The statistical nature of the efficient frontier: from confidence bands to the Resampled Efficient Frontier™
• Distorted risk and return: the impact of illiquidity, survivorship bias
• Robust portfolio construction approaches: minimizing regret

Exercise: Decision making under uncertainty in a scenario framework

Quantitative Portfolio Construction beyond Mean and Variance

• Overview of risk measures beyond volatility
  • Downside risk measurement: semi-variance, LPM, VaR/CVaR
  • Interim risk: drawdown measures (max drawdown, drawdown-at-risk)
• Higher moments
  • Skewness and kurtosis as tail risk characteristics
  • Their use in risk measurement and portfolio construction and estimation risk issues
• CVaR and LPM/UPM optimization: exact solutions and approximations
• Behavioural Portfolio Construction: Applying insights from Prospect Theory
• Random portfolios & convex hulls
• Understanding the advanced optimization algorithms: threshold accepting, simulated annealing, genetic optimizer
• Multi-criteria portfolio optimization: taking into account sustainability scores in the construction of ESG portfolio
• The use of AI (Artificial Intelligence), Machine Learning, Big Data Approaches in portfolio construction

Selected Topics in Quantitative Analysis

• Tail risk management
  • Understanding the normal distribution assumption
  • Non-normal distributions: Cornish-Fisher, NIG, normal mixtures
  • Tail risk contributions, risk budgeting with Modified VaR/CVaR
  • Modelling fat portfolio tails with elliptical distributions
• Overdiversification and diworsification
• Taking into accounting asymmetries in correlations: equities, bonds and gold

Exercise: Downside, upside, average and tail correlation measurement

Currency Management

• Risk, return and exposure analytics for unhedged and hedged currency exposures
• Approaches to currency management: active management versus hedging
• Optimal currency hedging: optimization approaches (1-stage, 2-stage) and practitioner approaches (strategic and tactical hedge quotas, currency overlay portfolios)

Exercise: 1-stage versus 2-stage optimization in currency management

Model Risk Management

• Forecast risk = model risk + parameter estimation risk
• Backtesting: in-sample versus out-of-sample, overfitting issues
• Elements of a buy-side model risk management framework