A workshop on statistical modeling with focus on time series analysis and grey-box modeling.

The purpose of the workshop is to give the participants an introduction to statistical modelling, especially to time series analysis using black-box (ARX/ARMAX) and grey-box models. The intent is that the workshop will be held in an interactive way, such that all participants independent of prior knowledge in the field, can gain new knowledge.

The program is as follows:

12:00: Arrival and sandwich lunch.
13:00: Lecture 1: Introduction of statistical modelling and time series modelling (Auto-correlation function, framework of models (regression, black box, grey-box), parameter estimation and uncertainty, statistical tests).
14:00: Exercise 1: ARX models for the heat dynamics of a test box (very small building). This exercise primarily aims at introducing the ARX models, which are extremely useful for forecasting and performance assessment.
18:00: Lecture 2: Maximum likelihood, model selection and model verification.
19:00: Dinner.

22:00: Possibility for staying over night at the location (free beds in the PowerFlexhouse).
08:00: Breakfast
08:30: Lecture 3: Grey-box modelling. The grey-box modelling framework provides an extremely strong setup for combining physics and statistics for modelling of dynamical systems, for example passive (mostly linear) and active (mostly non-linear) thermal systems providing the basis for MPC applications.
09:00: Exercise 2: Grey-box modelling of the heat dynamics of a building. The heat dynamics of a building are modeled with grey-box models and a model selection is carried out to find a suitable model structure.
12:00 Sandwich lunch and continuing on Exercise 2.
14:00 Goodbye.

The exercises will be "hands-on" using the participants laptops and carried out using the free software R (www.r-project.org) and CTSM-R (grey-box modelling package for R).