| 授業方針・テーマ |
This course is a comprehensive, practitioner-oriented guide to modern quantitative portfolio
management, with a strong focus on risk budgeting, which is the deliberate allocation and
control of risk across assets, strategies, or managers to improve risk-adjusted performance. The
course critically evaluates a wide range of techniques, from classic mean-variance optimization
to advanced methods that address real-world challenges such as estimation error, non-normal
returns, transaction costs, and out-of-sample performance. Key topics include portfolio resampling,
robust and Bayesian optimization (e.g., Black-Litterman), risk parity and equal risk
contribution approaches, diversification, core-satellite and benchmark-relative strategies, longshort
(e.g., 130/30 or 120/20) portfolios, scenario-based optimization, tail-risk hedging, and
dynamic tracking error management. |
習得できる知識・能力や授業の
目的・到達目標 |
The course aims to offer a balanced, evidence-based perspective and practical implementation
insights, bridging theory and application, making it valuable for portfolio managers,
quants, and advanced students seeking beyond basic frameworks. |
授業計画・内容
授業方法 |
1 Introduction
Mean–variance portfolio construction, its foundations and limitations; implied risk and return; risk budgeting versus optimization; covariance concepts.
2 Application
Practical applications including clustering, illiquid assets, and time-varying covariances in lifecycle investing.
3 Diversification
Approaches such as equal weighting, minimum variance, and risk parity, with key limitations including asset universe selection.
4 Risk Parity
Concept and implementation of risk parity, including factor and tail-risk extensions and links to asset pricing.
5 Non-Normality
Return non-normality; lower partial moments and comparison with variance-based approaches.
6 Resampling and Error
Estimation errors in returns and covariances; resampling techniques and their limitations.
7 Robust Optimization
Robust portfolio choice under uncertainty; constraints and regularization methods.
8 Bayesian Methods
Bayesian analysis; multivariate cases; Black–Litterman and incorporation of views.
9 Out-of-Sample Testing
Evaluation of methods under realistic conditions; constrained and unconstrained cases.
10 Transaction Costs
Effects of costs; turnover constraints and rebalancing strategies.
11 Core–Satellite
Active risk allocation and balance between alpha and beta.
12 Benchmark Optimization
Tracking error, benchmark-relative portfolios, and practical issues.
13 Relaxing Constraints
Long–short portfolios such as 120/20 and constraint effects.
14 Incentives
Performance fees, multi-period decisions, and dynamic tracking error.
15 Student presentations. |
| 授業外学習 |
1, September 12 (Saturday) 10:30-14:30: Q&A individual lecture (outline)
2. September 30 (Wednesday) 19:00 22:10: Q&A and individual lectures based on scoring and commentary (online) |
| テキスト・参考書等 |
Portfolio Construction and Risk Budgeting (5th Edition), Bernd Scherer, Risk Books (2015).
Algorithmic Finance: A Companion to Data Science, Christopher Hian Ann Ting, World-
Scientific (2022).
Machine Learning for Factor Investing–Python Version, Guillaume Coqueret and Tony Guida,
Chapman & Hall/CRC Financial Mathematics Series (2023). |
| 成績評価方法 |
Class Activity: 10%
Exercises: 65%
Presentation 25% |
質問受付方法
(オフィスアワー等) |
1. Office hours: Thursday 14:00~16;00 at Research Room 10 in person
2. Other week day by email online |
特記事項
(他の授業科目との関連性) |
|
| 備考 |
|
|
シラバス照会