| 授業方針・テーマ |
The industry landscape of investment, trading, and risk management has been revolutionized by computing technologies, data science, and financial engineering. The recent resurgence of AI and machine learning makes such a strong impact that the field of algorithmic quantitative finance will never be the same again.
To progress in tandem with the sea change in the industry, the topics covered in this courseinclude the applications of machine learning and data science to investment, alternative ETF construction, market microstructure, and algorithmic trading. |
習得できる知識・能力や授業の
目的・到達目標 |
In addition to mathematical modeling, an important component of this course is the practical aspect of computational implementations with statistical tests. Given that implementation and test procedures are involved, this course is anchored on algorithms and hands-on practicals (Python programming).
To facilitate effective learning, a mini project will be introduced to help you apply the algorithms for designing and evaluating exotic beta portfolio against TOPIX or other indexes. |
授業計画・内容
授業方法 |
1 Introduction
* Recent progress of Quantitative Finance
* Nikkei 225 Index: A Case Study
* Fundamental Indexation
* Python programming
2 Random Walks
* Random Walk Models
* Linear Scaling Law
* Variance Ratio Test
3 Event Study
* How does stock price react to new information?
* Abnormal return, cumulative abnormal return
* Long-short trading strategy in event study
4 Multi-Factor Analysis of Expected Returns
* Factor neutral portfolio construction
* Arbitrage pricing theory
* French’s data library
* ETFs
5 Factor Investing and Asset Pricing Anomalies
* Detecting anomalies
* Factors or characteristics?
* Momentum, timing, and ESG
* Links with machine learning
6 Portfolio Back-Testing I
* Setting the protocol
* Turning signals into portfolio weights
* Performance metrics
7 Portfolio Back-Testing II
* Common errors and issues
* Implications of non-stationarity: forecasting is hard
* A complete back-test
* Back-test overfitting
8 Causality and Non-Stationarity
* Granger causality
* Structural time series models
* Dealing with changing environments
* Links to online learning and homogeneous transfer function
9 Unsupervised Learning
* Principal Component Analysis (PCA) of yield curves
* Auto-encoders
* Clustering via k-means
10 Introduction to Market Microstructure
* Liquidity
* Limit order markets
* Bid-ask spread
* Tick data and trade direction
11 Mid-Term Summary
* Test
* Test review
12 Project
* Project assignment
* Python code examples
13 Penalized Regression and Hedging
* Penalized regression
* Hedging for minimum variance
* Predictive regressions
14 Validating and Tuning
* Learning metrics
* Validation
* Search for good hyperparameters
* Estimation and Forecasting
* Validation in backtests
15 Reinforcement Learning
* Theoretical layout
* Curse of dimensionality
* Policy gradient
* Simple Examples |
| 授業外学習 |
連続する２週間の集中講義を予定している．その間の週末の学習も必要になる． |
| テキスト・参考書等 |
・Advances in Financial Machine Learning, Marcos Lopez de Prado, John Wiley Sons (2018).
・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%
Test: 30%
Homework: 30%
Mini Project: 30% |
質問受付方法
(オフィスアワー等) |
http://cting.x10host.com/ |
特記事項
(他の授業科目との関連性) |
|
| 備考 |
Programming Language Python 3.12.x
「授業方針・テーマ」と「習得できる知識・能力や授業の目的・到達目標」に記載されている内容は、「Course Description」として書かれた文章を分割しただけなので、そのつもりでお読みください。 |
シラバス照会