Intermediate course, prerequisites: an introductory course in probability and statistics
Business Analytics refers to our ability to collect and use data to generate insights for fact-based decision-making. Every day our world is filled with new data, with every data input adding new information to the preceding ones. Google and Amazon, among many, are providing us with huge databases that record our preferences, a process made possible through the use of statistical learning. In this course we will explore challenges dealing with Big Data and learn about several statistical methods that are commonly used to investigate business-related problems. The course is designed for students with basic knowledge of statistics, and the content of the course will be of practical nature. It covers methods for data mining and business analytics and their usage in making strategic business decisions. It will concentrate on the modelling aspects of data mining and will provide students with a set of tools for better understanding key methods of, for example, data exploration, visualisation, classification, prediction, and clustering. The course starts with data visualization and getting to know features hidden in the data. Over time we will gain familiarity with traditional regression models and hypothesis testing and practice using them with real data. This introduction to traditional approaches will then lead to the discussion of more advanced methods such as, discriminant analysis, classification and clustering methods, which are useful in finding patterns hidden in the data. During the course, we deal with various types of data such as, categorical data, time series, text data, and network data, among others. The fundamentals of building suitable models are discussed. Illustrations are carried out using the statistical package R.
The course is designed as a series of lectures, computer exercises and assignments with reports. Grading is based on the assignments and a written exam.