Machine Learning in Economics and Business

About

Machine learning has become a dominant paradigm for data analysis. It is used across a vast variety of fields and applied problems, and it interests academics and practitioners alike. But what is machine learning really about? What does it mean that a machine “learns”? In what sense is machine learning the basis for artificial intelligence? What problems can it solve, and where are its limits? What are the most common algorithms of the machine learning paradigm? How does it relate to classical statistics and to econometrics? In this course, we address these questions. We do so in a very hands-on way. You learn the methods of machine learning by applying them directly during the course, using your laptop and programming in R. The emphasis is on both understanding the methods and on getting a feeling how they actually work in practice.

Course leader

Prof. Dr. Binswanger (University of St. Gallen, Switzerland)

Target group

This course is intended for students from the fields of economics or business with an interest in applying machine learning either for their doctoral thesis or in practical work thereafter. The course requires knowledge of the basic concepts of statistics, such as (multivariate) random variables, conditional probability, moments, maximum likelihood, the basics of multivariate OLS regression, as well as the basics of matrix algebra. For the handson parts of the course, you need some basic knowledge of the programming language R. You will be provided material for a “crash course” in R by June 2019.

Course aim

What does it mean that a machine “learns”? What problems can it solve, and where are its limits? What are the most common algorithms of the machine learning paradigm? How does it relate to classical statistics and to econometrics?

Fee info