MGT-302 / 2 credits

Teacher: Acerbi Carlo

Language: English

Remark: Une seule inscription à un cours SHS+MGT autorisée. En cas d'inscriptions multiples elles seront toutes supprimées sans notification

## Summary

This course focuses on methods and algorithms needed to apply machine learning with an emphasis on applications in business analytics

## Content

The following topics will be covered in the course:

1. Supervised learning
- Linear Regression
- Multiclass Classification
- K-NN
- Support Vector Machines
- Decision Tree and Random Forest

2. Unsupervised learning
- Regularization and Model Selection
- Cross Validation
- PCA

3. Deep Learning
-Deep Neural Networks
-Back propagation

4. Graphical models

-Bayesian networks
-Inference and structure learning

5. Causal inference in time series
- Granger causality
- Directed information Graphs

6. Quantitative Risk Management
- Risk Measures: Value at Risk and Expected Shortfall
- Statistical Estimation and Risk Measurement

7. Statistical Learning for Finance
- Shrinkage, Ridge Regression, LASSO and Dimension Reduction
- Predicting Financial Returns

## Keywords

machine learning, causal inference, time series, quantitative risk management

## Required courses

A course in basic probability theory
A course in basic linear algebra
Calculus
Familiarity with Python or Matlab

## Important concepts to start the course

Students should be familiar with basic concepts of probability theory, calculus, linear algebra, and
programming.

## Learning Outcomes

By the end of the course, the student must be able to:

• Formulate supervised and unsupervised learning problems and apply it to data

## Transversal skills

• Assess one's own level of skill acquisition, and plan their on-going learning goals.

## Teaching methods

Formal teaching interlaced with practical exercices.

## Expected student activities

Attending lectures and working on homework and projects.

## Assessment methods

Three homeworks (33.33333333% each)

## Supervision

 Office hours Yes Assistants Yes Forum No

## In the programs

• Semester: Spring
• Number of places: 80
• Exam form: During the semester (summer session)
• Subject examined: Introduction to data driven business analytics
• Lecture: 2 Hour(s) per week x 14 weeks
• Type: mandatory

## Related courses

Results from graphsearch.epfl.ch.