OT-SC-WS-02 | Quantitative analyses for data science
Prof. Dr. Thorsten Dickhaus
Proficiency in (mathematically grounded) quantitative data analysis is key to many modern applications in data science. Understanding the basic underlying principles helps to interpret data analysis results, even if one does not analyze the data by oneself.
Basic notions of mathematical and applied statistics are presented. Some prototypical statistical models, in particular regression models and time series models, are treated in more detail. Major topics are point estimation, confidence estimation, testing, and prediction. At some occasions, connections to statistical (machine) learning are drawn. The course consists of lectures and practical hands-on sessions.
Principles of decision making under uncertainty</li> <li>Statistical data modeling</li> <li>Statistical data analysis</li> <li>Interpretation of statistical data analysis results
- Basic mathematical education (maps, matrices, taking derivatives, solving integrals, matrix-vector multiplication, …)
- Knowledge in basic probability theory (probability spaces, random variables, random vectors, probability distributions, central limit theorem, …)
- Own PC, laptop with R software installed (R + R-Studio)
- Paper and pens
- For online format a second screen might be beneficial
- Robert W. Keener (2010): Theoretical Statistics. Topics for a Core Course. Springer.