Learning outcomes:
The students understand the importance of a statistical description of data sets and can apply the concepts of moments of a distribution (average, median, variance, skewness) on arbitrary data sets. They can perform the significance test to identify influence parameters that show a strong impact on the data sets. Furthermore, they are able to formulate and test different hypotheses and to pro-vide a significance level for the acceptance or rejection of a hypothesis. The students will write a numerical code to perform such tests and apply it to the formulation of mathematical models for the description of data sets. They will also understand the basic principles of the Design of Experi-ments (DoE) methods, as sampling of parameter spaces in contrast to high-throughput generation of data, and formulation of empirical models. In simple numerical examples they will apply this knowledge to independently solve given problems.
Subject aims
- Introduction into statistical description of data sets
- Reliability of data and sources of error
- Fitting and smoothing of data
- Formulation of hypotheses and significance tests
- Design of experiments (DoE)
- Different methods for sampling of parameter spaces
- High-throughput data generation, management and storage
- Identification of true effects and interactions
- Quantitative description of data with empirical models
- Hands-on exercises on data analysis and DoE
- Kursleiter/in: Irina Roslyakova