Statistical discrimination without knowing statistics : can we blame social interactions ?

Emily Tanimura - Université Paris-1 Panthéon-Sorbonne (CES)
vendredi 31 janvier 2014

Statistical discrimination occurs when an individual receives a treatment that is based on the characteristic of the group he belongs to (for example a high performing individual is not hired because he belongs to an on average low performing group). Usually the underlying assumption is that the decision maker knows the characteristics of the group but not the characteristics of the individual. In this work we show that outcomes perfectly similar to those that occur with statistical discrimination can occur under an entirely different behavioral assumption : decision makers observe only individual characteristics and ignore the distribution of characteristics in the population but are sensitive to social influence. We study a dynamic setting of repeated decisions and compare the outcomes with and without social interaction between decision makers.


Agenda

<<

2017

>>

<<

Mai

>>

Aujourd'hui

LuMaMeJeVeSaDi
1234567
891011121314
15161718192021
22232425262728
2930311234

Annonces

ESANN 2016 : European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning


STATLEARN 2016


ICOR 2016