Financial big data
FIN-525 / 3 credits
Teacher: Challet Damien
Language: English
Withdrawal: It is not allowed to withdraw from this subject after the registration deadline.
Remark: MA3 only
Summary
The course introduces modern methods to acquire, clean, and analyze large quantities of financial data efficiently. The second part expands on how to apply these techniques and robust statistics to financial analysis, in particular to intraday data and investment strategies.
Content
Big Data
- The future of storage, computing power, efficiency
- Financial data sources and acquisition
- Data cleaning and formatting
- Efficient visualization techniques
- Robust estimators for financial data
- Multicore/GPU and cluster computing
- Out-of-core data analysis
Application to financial data
- Intraday tick-by-tick data
- Dimensionality reduction
Correlation matrix cleaning with random matrix theory
Clustering of assets and days - Brute-force trading strategy design and selection
Non-stationary predictions
Best and worst practices
Keywords
Big Data, stylized facts, data wrangling, dimension reduction, tick-by-tick data, trading strategy, strategy selection.
Learning Prerequisites
Required courses
- Very good programming skills (required) and a first experience with R or/and Python.
- Good knowledge of the probability and statistics concepts taught in the first (two) year(s) at EPFL. This includes the Central Limit Theorem and its important applications in statistics.
Recommended courses
- Advanced statistics
- Econometrics
- Investments
- Programming with R, or Python.
Important concepts to start the course
See above
Learning Outcomes
By the end of the course, the student must be able to:
- Choose appropriate methods and tools to manipulate and analyze complex financial data.
- Conduct efficient data cleaning.
- Implement financial big data analysis using R and Python
- Implement proper computationally intensive strategy backtests
- Plan computing resource usage time
- Infer financial measurables with robust estimates
Transversal skills
- Collect data.
- Write a scientific or technical report.
- Demonstrate a capacity for creativity.
- Access and evaluate appropriate sources of information.
- Continue to work through difficulties or initial failure to find optimal solutions.
Teaching methods
3 hours of ex-cathedra lectures and supervised applications for 14 weeks
Expected student activities
- Actively participating at lectures
- Completing theoretical and practical exercices during the lectures.
- Writing up of a report which demonstrates the ability to apply the concepts and tools introduced in this course.
Assessment methods
- Group projects 100%
Supervision
| Assistants | Yes |
| Others |
Resources
Bibliography
Ressources en bibliothèque
- Empirical properties of asset returns: stylized facts and statistical issues / Cont
- An Introduction to Statistical Learning [2n edition, 2022] / James, Witten, Hastie, Tibshirani
- Python for Finance / Hilpisch
- Analysis of Financial Times Series [3rd edition 2010] / Tsay
Références suggérées par la bibliothèque
Moodle Link
In the programs
- Semester: Fall
- Exam form: During the semester (winter session)
- Subject examined: Financial big data
- Lecture: 3 Hour(s) per week x 14 weeks
- Type: optional
- Semester: Fall
- Exam form: During the semester (winter session)
- Subject examined: Financial big data
- Lecture: 3 Hour(s) per week x 14 weeks
- Type: optional
- Semester: Fall
- Exam form: During the semester (winter session)
- Subject examined: Financial big data
- Lecture: 3 Hour(s) per week x 14 weeks
- Type: optional
- Semester: Fall
- Exam form: During the semester (winter session)
- Subject examined: Financial big data
- Lecture: 3 Hour(s) per week x 14 weeks
- Type: optional
- Semester: Fall
- Exam form: During the semester (winter session)
- Subject examined: Financial big data
- Lecture: 3 Hour(s) per week x 14 weeks
- Type: optional