Skill course: Applied Statistics and Data Analysis
Typical standard statistics curricula taught in Undergraduate courses strongly build up upon Gaussian statistical theory, where a number of clauses and statistical applied tests and procedures can be derived from the properties of the Gaussian distribution. However through development of theory and almost as importantly computational power a number of techniques have been developed that relax Gaussian assumptions, are more suitable for real world problems, often appear to be more intuitive and are increasingly applied in geoscientific data analysis. Moreover, graphical and data-adaptive data analysis approaches will be taught. Finally, techniques for model evaluation and calibration will be introduced. This course will give an introduction to these techniques, based on lectures (mostly morning) and exercises (mostly afternoon)
Date & Place
March 14-18 2011
Seminar room B0.002
What to prepare before March 14
- Please make sure you have a running copy of R on your laptop.
- You can download the most recent version here: http://www.r-project.org/
- RStudio is a new open-source integrated development environment that runs on all platforms. It nicely combines console, script editor, working directory, plots etc. into a an uncluttered layout that you can easily navigate. You need to have R installed before you can use RStudio as a development environment.
- To brush up your R-skills, look here.
- Also, check that you can access the BGC user WLAN with your laptop. Please see your friendly IT-Team in time if you have difficulties to set this up.
Please do these checks in the week prior to the course so we don't loose precious time in class.
Slides & Data
Data for the course can be found here:
- ftp ftp://ftp.bgc-jena.mpg.de/pub/outgoing/mreichstein/4IMPRS-gBGC/, organized by days (NEW: R-scrips of Thursday uploaded!)
- and possibly here on rhea: tmp/_IMPRS_courses/Statistics2011
- intro by Markus Reichstein
- slides by Jens Schumacher (4 slides on one page)
- slides by Jens Schumacher (1 slide = 1 page)
- slides for practical by Jens Schumacher
- R scripts by Jens Schumacher
- Example Data Jens Schumacher
- Slides Friday by Thomas Wutzler on nonlinear parameter estimation
- short reference card for R
- introduction to R
Useful background information
- Zuur et al. (2009) Mixed Effects Models and Extensions in Ecology with R.
To guarantee a good supervision, the number of places is limited. Right now all places are filled and the registration is closed. Please contact Anna in case there is a problem.
At the bottom of this page is a list of people who have already registered.
Legend T = theory, M = methods, d = demonstration
|Mon||General statistical/stochastic concepts and advanced descriptive statistics, ANOVA and General Linear Models||Jens Schumacher|
Introduction to general statistical concepts
Simple and multiple linear regression
One- and two-way ANOVA
|p.m. (I)||D||Software for data analysis|
|p.m. (II)||M||Tutorial on descriptive and graphical presentation of data|
|Tue||General statistical/stochastic concepts and advanced descriptive statistics, ANOVA and General Linear Models||Jens Schumacher|
|a.m.||T||General linear model|
Regression models for temporally dependent data
Regression models for spatially dependent data
|p.m.||M||Computer tutorial on regression approaches|
|Wed||Explorative and multivariate data analysis||Miguel Mahecha|
|a.m.||T||Intro to multivariate data sets|
Data adaptive prediction (by Antje Moffat and Martin Jung)
|Thu||Modelling and Model evaluation||Markus Reichstein|
|a.m.||T||Conceptual introduction to model evaluation|
Simple Model evaluation metrics
Advanced model evaluation metrics (by Miguel Mahecha)
Bootstrapping: a teaser
|14:00-15:00||break for MPI-BGC colloquium on tree rings by Rolf Siegwolf|
|p.m.||M||Computer tutorial on model evaluation metrics|
Computer tutorial on uncertainty analysis
|Fri||Non-linear model parameter estimation I: LSQ techniques, Bayesian approaches||Thomas Wutzler|
|a.m.||T||Overview of parameter estimation|
Parameter Estimation: Likelihood principles
Parameter Estimation: Bayesian principles
|p.m.||M||Computer tutorial on model parameter estimation|