An introduction to modern statistical and machine learning with applications in R. Topics to be covered include

• regularized regression (lasso and elastic net),
• model assessment and selection (cross-validation and bootstrap),
• regression and classification trees,
• ensemble methods (bagging, random forests, boosting),
• deep learning with neural networks.

After completing this course the student will have the ability to:

• describe and apply key methods of statistical and machine learning including regularized regression, regression and classification trees, ensemble methods and deep learning with neural networks;
• interpret estimation results and perform model assessment and selection, e.g., using cross-validation and bootstrapping.

Moreover, after completing this course the student will have the ability to:

• adequately communicate the results of fitting a suitable statistical and machine learning model to data;
• critically assess the application of statistical and machine learning methods for data analysis.

In addition, the student will be able to:

• use R to perform data analysis using statistical and machine learning methods.

Full attendance is compulsory. This means that students should attend at least 80% of all lectures, at most one lecture can be missed.

Students need to attend the exercise classes in order to be able to tick the assignments and have them counted. There are five exercise classes. For each exercise class students can obtain 10 points; the best four out of the five exercise classes will be used for the final score. This means that missing one exercise class is possible while still obtaining full points for the assignments part.

Students need to present their course project and write the exam in person. Attendance is thus compulsory for these units to obtain points for the final score.

The course is taught as a lecture combined with exercise classes where assignments and a course project are presented by students. While lectures are taught with all students enrolled in this course subject together, there are four courses in parallel for the exercises classes where the assignments and the course projects are presented by students split into smaller groups.

For the assignments in the exercise classes, students prepare assignments at home and indicate by completing an assignment on Canvas WU with common deadline the latest one hour before the exercise classes start which assignments they prepared and are able to present in class. During the exercise classes, students who ticked that they prepared the assignment via Canvas WU are randomly selected to present their solution of the assignment. For the presentation to be deemed suitable, the student must have a complete solution proposal where all points of the assignment are fully addressed. Incomplete solution proposals lead to all exercises ticked for this exercise class being canceled for this student.

The course projects are done in groups also separately for each of the parallel courses. Students forming a group need to be enrolled in the same course and have a time slot assigned for their presentation from the time slots allocated for this course for project presentations.

In combination with the lecture, the assignments will help students to consolidate and expand their knowledge and understanding of the statistical and machine learning methods covered in class as well as model assessment and selection approaches by developing solutions to theoretical and applied problems.

The course project will help students to gain some experience in applied data analysis using statistical and machine learning methods by applying several statistical and machine learning methods in combination with model assessment and selection methods to arrive at suitable model. Students will work on the course project in groups and present their results in class.

The assessment is based on three evaluation criteria with the following weights:

• 40% assignments (individual, presented in the exercise classes; best 4 out of 5 classes count; points obtained depend on the number of assignments ticked as well as the presentation performance).
• 20% course project (group work, presented in the exercise classes).
• 40% written final exam (individual).

Grading key: Unsatisfactory: <50%; Adequate: <65%; Satisfactory: <80%; Good: <90%; Very good: <=100%.

Please log in with your WU account to use all functionalities of read!t. For off-campus access to our licensed electronic resources, remember to activate your VPN connection connection. In case you encounter any technical problems or have questions regarding read!t, please feel free to contact the library at readinglists@wu.ac.at.