363-1098-00L Business Analytics
| Semester | Spring Semester 2019 |
| Lecturers | S. Feuerriegel |
| Periodicity | yearly recurring course |
| Language of instruction | English |
| Comment | Students from the MAS MTEC are not applicable for this course and are kindly asked to enroll in the course "Executive Business Analytics (365-1120-00L)" instead. |
| Abstract | Prior to the start of the Information Age in the late 20th century, companies were forced to collect data from non-automated sources manually. Companies back then lacked the computing capabilities necessary for data to be analyzed, and as a result, decisions primarily originated not from knowledge but from intuition. |
| Learning objective | With the emergence of ubiquitous computing technology, company decisions nowadays rely strongly on computer-aided “Business Analytics”. Business analytics refers to technologies that target how business information (or sometimes information in general) is collected, analyzed and presented. Combining these features results in software serving the purpose of providing better decision support for individuals, businesses and organizations. This course will teach what distinguishes the varying capabilities across business analytics – namely the underlying methods. Participants will learn different strategies for data collection, data analysis, and data visualization. Sample approaches include dimension reduction of big data, data visualization, model selection, clustering and forecasting. In particular, the course will teach the following themes: • Forecasting: How can historical values be used to make predictions of future developments ahead of time? How can firms utilize unstructured data to facilitate the predictive performance? What are metrics to evaluate the performance of predictions? • Data analysis: How can one derive explanatory power in order to study the response to an input? • Clustering: How can businesses group consumers into distinct categories according to their purchase behavior? • Dimension reduction: How can businesses simplify a large amount of indicators into a smaller subset with similar characteristics? During the exercise, individual assignments will consist of a specific problem from business analytics. Each participant will be provided with a dataset to which a certain method should be applied to using the statistics software R. |
| Content | Content: 1. Motivation and terminology 2. Business and data understanding a. Data management and strategy b. Data mining processes 3. Data preparation for big data a. Software and tools b. Knowledge representation and storage c. Information preprocessing 4. Explanatory modeling 5. Predictive modeling a. Classification b. Variable selection c. Handling non-linearities d. Ensemble learning e. Unsupervised learning f. Working with unstructured data 6. Managerial implications |
| Literature | James, Witten, Hastie & Tibshirani (2013): An Introduction to Statistical Learning: With Applications in R. Springer. Sharda, Delen & Turban (2014): Business Intelligence: A Managerial Perspective on Analytics. Pearson. |