Infra Informatics - Statistical Research Methods and Data Analysis, 2.0 credits

Entry requirements

Admitted as a doctoral student.

Specific information

The course aims to

• give an overview of different statistical learning methods
• discuss methodology, tools, and practices for applied research in statistical learning
• present data-driven research projects within Infra Informatics.

Learning outcomes

After completing the course, participants should be able to:

• describe and categorize various statistical learning methods,
• describe and discuss the suitability, possibilities, and limitations of different statistical methods in relation to certain problem settings,
• identify, select, and plan the necessary steps for conducting a successful data-driven research project,
• describe and discuss data-driven research as well as the use of different statistical methods within Infra Informatics.

Contents

The following topics will be covered (in varying depth):

• Research methodology in data-driven research projects
• Overview of statistical methods and machine learning
• Estimation, prediction and inference
• Overfitting and bias-variance trade-off
• Supervised learning: different methods for regression and classification
• Unsupervised learning: clustering, dimensionality reduction and density estimation
• Resampling and evaluation methods
• Hypothesis testing
• Tools and programming languages for statistical data processing

Educational methods

The course is held during VT1 each year. The schedule consists of

• A half-day startup meeting, presenting the common framework for the whole course and first lectures on the course topics.
• Two full day seminar days, with lectures on the course topics, a lab session and presentations by senior researchers at KTS about projects where statistical methods have been applied.
• An individual homework assignment.
• A half-day final meeting, with student presentations and discussions.

Examination

Responsible for the Statistical Research Methods and Data Analysis course is David Gundlegård.

The examination for the course consists of:

• Mandatory participation in the four seminar occasions
• Conducting, documenting, and presenting an individual assignment

Grading

General information

The course is mandatory for all doctoral students in Infra Informatics. It is also open to doctoral students in other fields.