May. 19 2009 2:00PM

May. 19 2009 3:30PM

Understanding the dynamics of networked systems in business – in particular the interplay between the performance of an individual node and of the entire network – becomes a critical skill. Replacing mental models with a formal, quantitative model can improve such understanding and ultimately allow for systematic network optimization. To this end, the combination of stochastic system dynamics modeling of individual nodes with probabilistic graphical modeling of a network configuration will be proposed.
The seminar will discuss the economic substance and mathematical representation of node-to-node bonds. A general Bayesian solution of estimating unknown state and parameter values in the resulting model will be formulated. And finally, Markov chain Monte Carlo implementation of the proposed solution will be discussed.

Pravidelný seminář Rozhodování a řízení za neurčitosti. (Contact person: Miroslav Kárný)