Tid: 24 februari 2015 kl 10.00-Seminarierummet 3721, Institutionen för matematik, KTH, Lindstedts vägen 25, plan 7. Karta!
Föredragshållare: Jonas Hallgren, KTH
Titel: Licentiat seminarium: Continuous time Graphical Models and Decomposition Sampling
Huvudhandledare: Professor Timo Koski
Diskutant: Professor Błazej Miasojedow, Institute of Applied Mathematics and Mechanics, University of Warsaw
Abstract: Two topics in temporal graphical probabilistic models are studied. The topics are treated in separate papers, both with applica- tions in finance.
The first paper studies inference in dynamic Bayesian networks using Monte Carlo methods. A new method for sampling random variables is proposed. The method divides the sample space into subspaces. This allows the sampling to be done in parallel with independent and distinct sampling methods on the subspaces. The methodology is demonstrated on a volatility model and some toy examples with promising results. The second paper treats probabilistic graphical models in continuous time |a class of models with the ability to express causality. Tools for inference in these models are developed and employed in the design of a causality measure. The framework is used to analyze tick-by-tick data from the foreign exchange market.
|Sidansvarig: Filip Lindskog