Leibniz Network "Mathematical Modeling and Simulation"
MMS Summer School 2018
Statistical Modeling and Data Analysis
September 3, 2018 - September 7, 2018
Experimental or observational data of high or infinite dimensionality are getting common in institutes of all sections of the Leibniz Association. This creates an increasing demand for adequate modern data analysis techniques. At the same time reproducibility of experiments and their statistical analyses lead to new requirements for good scientific practice and requests for open source and open science.
We will address both topics with the first PhD Summer School of the Leibniz Network “Mathematical Modeling and Simulation” in a way that provides knowledge transfer from mathematical and applied statistics into various scientific communities and helps to develop skills in R programming, statistical modeling and reproducible data analysis. The School will be problem oriented. Participants are asked to provide their data and problems in advance (see below).
Introductory morning sessions are planned to cover the following topics:Complex 1: R programming and good scientific practice
- R and R programming,
- Version control
- Dynamic documents and reproducible research with R using knitr and LaTeX
- Open source and open science, requirements for reproducible research and publications
Complex 2: Modeling high-dimensional data (will be adjusted according to the particular problems provided by the participants)
- Multiple testing using False Discovery Rates (FDR)
- Dimension reduction, low rank approximation, sparsity regularization
- Reproducible variable selection in regression and classification
Complex 3: Statistical methods for functional data (will be adjusted according to the particular problems provided by the participants)
- Curves (growth curves, spectra, ...) and images as covariates or response
- Mean, median and covariances of functional data
- Dimension reduction, regression, classification
- Registering functional data (warping) and shapes
Afternoon sessions in smaller groups will be used to discuss selected submitted problems and to model and analyze the corresponding data. Data analysis will be performed using the R Language and Environment for Statistical Computing and Graphics. Participants are asked to bring their own laptop with R and preferably R-studio installed.
We plan the Summer School to be problem oriented. Ideally, we would center the afternoon discussion and modeling sessions around problems participants are actually working at. We therefore ask participants to provide information on the topic and field of their PhD thesis as well as for introductory references on experiments and problems that they are interested in. If possible we would like to ask participants to provide (anonymized) example data for and a description of their data analysis problem, preferably no later than end of June.
- Dr. Clara Happ (LMU Munich, AG Biostatistics): Functional Data Analysis
- Dr. Joerg Polzehl (WIAS): Modeling High-dimensional Data; person in charge of the scientific program
- Dr. Heidi Seibold (LMU Munich, Institute for Medical Information Processing, Biometry, and Epidemiology): R, Open Science, Reproducible Research
- Almond Stöcker (LMU Munich, AG Biostatistics): Functional Data Analysis
- Dr. Alexandra Suvorikova (WIAS): Mathematical Statistics
Participation including housing and meals is free for PhD students from network member institutions. Travel expenses for the journey to Oberwolfach have to be covered by the hosting Leibniz institutions.
Registration deadline was June 15, 2018.