English   Danish

2024/2025  KAN-CMECV1247U  Predictive Modeling and Machine Learning

English Title
Predictive Modeling and Machine Learning

Course information
Language English
Course ECTS 7.5 ECTS
Type Elective
Level Full Degree Master
Duration One Semester
Start time of the course Spring
Timetable Course schedule will be posted at calendar.cbs.dk
Max. participants 80
Study board
Study Board for HA/cand.merc. i erhvervsøkonomi og matematik, MSc
Course coordinator
  • Jonas Striaukas - Department of Finance (FI)
Main academic disciplines
  • Mathematics
  • Statistics and quantitative methods
Teaching methods
  • Blended learning
Last updated on 25-01-2024

Relevant links
Learning objectives
At the end of the course, students are expected to be able to
  • explain and justify the main methods within machine learning
  • apply the most important machine-learning methods in specific problems using relevant statistical software
  • rigorously describe the strengths and weaknesses of the different methods
  • choose between different methods in a concrete data problem and justify this choice
  • communicate the results of an analysis performed with machine-learning methods
Course prerequisites
The course is aimed at students whose prerequisites correspond to an HA (math.). The teaching is in English, so adequate English is a prerequisite as well.
Examination
Predictive Modeling og Machine Learning:
Exam ECTS 7,5
Examination form Oral exam
Individual or group exam Individual exam
Duration 30 min. per student, including examiners' discussion of grade, and informing plus explaining the grade
Preparation time No preparation
Grading scale 7-point grading scale
Examiner(s) Internal examiner and second internal examiner
Exam period Summer
Make-up exam/re-exam
Same examination form as the ordinary exam
Description of the exam procedure

At the oral exam, a topic is drawn (randomly) from a list of known topics. The student can choose to let a presentation of a solution of a corresponding exercise sheet be part of the oral presentation.

 

The exam is in English.
Course content, structure and pedagogical approach

The course provides an introduction to the topics that are often referred to as artificial intelligence, data mining, and machine learning. During the course, the focus is going to be on the understanding of the methods and their theoretical foundation, as well as on concrete data applications of the various methods.

 

Specifically, the course covers the following topics:

 

  • Non-parametric methods for function estimation: Smoothing, splines, regularization, additive models
  • Penalized regression and classification: Properties, model selection, and inference
  • Resampling methods, including cross-validation
  • Tree-based models: Bagging, random forests, and boosting
  • Neural networks, including structural networks such as recurrent and convolutional neural networks
  • Selected topics on causal machine learning
Description of the teaching methods
The course consists of lectures and exercise classes. During the exercise classes, work is done on exam-relevant exercises with the help of the course's teachers. These assignments represent an exam topic — typically one or more machine-learning methods. The solution to these tasks can be discussed with the teachers during the exercise lessons, but there will be no opportunity for further feedback until the exam.
Feedback during the teaching period
During the exercise classes, exercises are worked on in small groups in a constructive dialogue with the teachers.

The lectures contain small quizzes and assignments, where the answers are discussed together.

Detailed reviews of exam-relevant examples are presented at the lectures and uploaded on the course's website.
Student workload
Lectures 24 hours
Exercise classes 12 hours
Examination 30 hours
Preparation 140 hours
Expected literature

The main coursebook:
Fan, Li, Zhang, and Zou (2020). Statistical foundations of data science. Chapman and Hall/CRC.

 

Supplementary literature:

 

Predictive machine learning:

Hastie, Tibshirani og Friedman (2009). The Elements of Statistical Learning: Data Mining, Inference and Prediction. Anden udgave, Springer.

▶ online copy:   https:/​​/​​hastie.su.domains/​​ElemStatLearn/​​printings/​​ESLII_print12_toc.pdf

 

Causal machine learning:

Huber (2013). Causal Analysis: Impact Evaluation and Causal Machine Learning with Applications in R. MIT Press.

 

Neural networks:

Aggarwal (2018). Neural Networks and Deep Learning, Springer.
Last updated on 25-01-2024