News

[ News List | AS seminars | CSKI seminars soon | CSKI seminars ]

Scientific Group "Decision-Making and Control under Uncertainty" (DCU)


List of the seminars:


16.2.2010  
Petr Volf:
On probabilistic models for the score in sport matches

Details of the seminars:


      Petr Volf:

On probabilistic models for the score in sport matches
Tuesday, 16. 2. 2010, 1400
room 474, UTIA ASCR

       A basic model for final score in a sport game (as for instance football or hockey) assumes that the numbers of goals scored by home and away teams are independent Poisson variables. Parameters can depend on factors characterizing conditions of the match, strength of teams, etc. Evident weak aspects of such a model (independence, absence of time factor) led to improvements proposed by various authors.
       I shall review briefly some of them. Then I shall introduce a model in which the goals are a realization of two dependent random point processes. It is assumed that the scoring intensity of each team has several components depending on time and again on factors describing the teams and other conditions of the match. This dependence is modelled with the aid of a semiparametric multiplicative regression model of intensity. A method of model evaluation is presented and demonstrated on small real data from the football World Championship 2006.
       The model is also used for random generation of artificial results.

Doc. Petr Volf, CSc.
Dep. of Stochastic Informatics


      ▲ List