Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

SOE: Fachverband Physik sozio-ökonomischer Systeme

SOE 21: Economic Models and Evolutionary Game Theory II

SOE 21.2: Vortrag

Donnerstag, 29. März 2012, 15:45–16:00, H 0110

An approach to stochastic social modeling: the second moment variables — •Felipe Lara-Rosano — Universidad Nacional Autonoma de Mexico, Mexiko Stadt, Mexiko

In this paper we introduce the second moment probabilistic approach as a way to model uncertainty in social phenomena like risk management. In order to express uncertainty in second moment terms, we will adopt a subjective or Bayesian probabilistic approach.

A second moment random variable (vector), SMRV, is a random variable (vector) for which the mean value (vector) and the variance (covariance matrix) have been assigned. Different random variables having different probability distributions but identical mean value and variance are then identical as second moment random variables. Also a SMRV defines a random vector only to within the class of random vectors that have the given mean vector and covariance matrix. This approximation is often sufficient for social and management applications. In fact in economic forecasting and decision making it is usual to consider only expected values as a first approximation.

Considering the fact that most social data are collected in a discrete way, for instance, from year to year, we will refer in this paper to time series analysis. We will show how a second moment markovian sequence can be used to model vector time series and how it can be manipulated as the state vector of a linear discrete dynamic system, offering a wide field of applications.