MBM437 Introduction to Data Science for Marketing

• Topics

  In this course, we will follow closely the structure of our textbook (i.e., Imai 2018) to present students with the important topics in the field of data science for marketing. The tentative list is below and subject to change:

  1.Causality

  While causality is one of the most crucial topics in science in general and data science in particular, it is also a difficult topic to study for many people. In this session, we will discuss important conditions to make a (valid) causal claim and what to do when those conditions are not met by the data (e.g., selection biases in experiment-based vs. observational data).

  2. Measurement

  Another fundamental topic in data analysis is measurement. Here we will talk about different measurement issues that can occur with a given marketing data set such as reliability and validity. We will also discuss the importance of sampling and the representativeness of our sample.

  3. Prediction

  It is common for marketing researchers or data analysts to predict future customer behaviors, the firm’s market share, or the future conditions of the market. In this session, we will discuss possible modeling approaches to do prediction (e.g., linear regression models) and how to interpret the results.

  4. Discovery

  Data exploration is also another important task when doing data analysis, especially when the data is not in the usual format. In this session, students will be introduced to different types of data (e.g., Twitter data) and how to extract insights as well as to discover underlying patterns from them.

  5. Probability and uncertainty

  There is a lot of uncertainty when we work with data and depending on our purposes, we often need to calculate certain probability conditional upon a given list of assumptions. With this topic, our aim is to

  introduce students with the relevant concepts about probability and uncertainty, in addition to the major differences between the two popular (and dominant) perspectives: frequentist and Bayesian.

  Along with the above topics, students will also get to be introduced to R (and RStudio) and how to use this software to assist them in all the data science assignments.

• Learning outcome

  Learning outcomes

  Knowledge

  After completing the course, students should know:

  • the basics of data science for marketing and relevant concepts
  • the different characteristics of different types of data
  • the connections between practical business problems and required data
  • a number of data science methods to analyze data and how to apply it in marketing tasks

  Skills

  After completing the course, students should be able to:

  • use R for data cleaning, manipulation, preparation, and visualization
  • work with raw data of different types for further analyses
  • perform a number of data science methods to analyze data
  • understand the benefits and limitations of a selected number of data science methods for marketing tasks

  General Competence

  After completing the course, students should be able to:

  • develop a solid understanding of the study topics as well as the applications of data science in marketing
  • reflect on the benefits and limitations of the discussed data science methods when working with (marketing) data
  • be ready for more advanced courses in marketing and customer analytics

• Teaching

  Teaching will contain synchronous teaching activities such as in-class lectures and lab sessions, as well as asynchronous learning activities such as external videos, reading materials, etc.

• Recommended prerequisites

  Knowledge of statistics would be an advantage.

• Required prerequisites

  None

• Compulsory Activity

  There will be 6-7 class small exercises throughout the course where students will work (individually or in groups, maximum of 4 students per group) on some data science assignments. You are required to pass at least 4 of them to get course approval.

• Assessment

  Term paper that can be done individually or in groups (maximum of 4 students per group), which is 100% of the final grade in the course.

• Grading Scale

  Grading scale: A-F

• Computer tools

  R, RStudio

• Literature

  Imai, K. (2018). Quantitative social science: an introduction. Princeton University Press.

  And a list of scientific articles.