Data, Models and Decisions in Business Analytics


In today’s world, managerial decisions are increasingly based on data-driven models and analysis using statistical and optimisation methods that have dramatically changed the way businesses operate in most domains including service operations, marketing, transportation, and finance.

The main objectives of this course are the following:

  • Introduce fundamental techniques towards a principled approach for data-driven decision-making.
  • Quantitative modelling of dynamic nature of decision problems using historical data, and
  • Learn various approaches for decision-making in the face of uncertainty

Topics covered include probability, statistics, regression, stochastic modelling, and linear, nonlinear and discrete optimisation. Most of the topics will be presented in the context of practical business applications to illustrate its usefulness in practice.

Associated Programmes: Business Analytics MicroMasters® Programme

Business Analytics MicroMasters® Programme
Columbia’s MicroMasters® programme in Business Analytics will empower learners with the skills, insights and understanding to improve business performance using data, statistical and quantitative analysis, and explanatory and predictive modelling to help make actionable decisions.

Analytics in Python
Learn the fundamental of programming in Python and develop the ability to analyse data and make data-driven decisions.
View the course

Data, Models and Decisions in Business Analytics
Learn fundamental tools and techniques for using data towards making business decisions in the face of uncertainty.
View the course

Marketing Analytics
Develop quantitative models that leverage business data to forecast sales and support important marketing decisions.
View the course

Demand and Supply Analytics
Learn how to use data to develop insights and predictive capabilities to make better business decisions.
View the course

Prerequisites: Undergraduate probability, statistics and linear algebra. Students should have working knowledge of Python and familiarity with basic programming concepts in some procedural programming language.

edX® is a registered trademark of edX® Inc.  All Rights Reserved.


  • Programme duration
    12 weeks
  • Estimated effort
    8 to 10 hours per week
  • Fee
  • Institution
  • Language
Start Dates
  • Fundamental concepts from probability, statistics, stochastic modelling, and optimisation to develop systematic frameworks for decision-making in a dynamic setting
  • How to use historical data to learn the underlying model and pattern
  • Optimisation methods and software to solve decision problems under uncertainty in business applications

Course Syllabus
Introduction to Probability: Random variables; Normal, Binomial, Exponential distributions; applications
Estimation: sampling; confidence intervals; hypothesis testing
Regression: linear regression; dummy variables; applications
Linear Optimisation: Non-linear optimisation; Discrete Optimisation; applications
Dynamic Optimisation: decision trees