eduzhai > Applied Sciences > Computer Science >

Aggregative Efficiency of Bayesian Learning in Networks

  • KanKan
  • (0) Download
  • 20210425
  • Save

... pages left unread,continue reading

Document pages: 40 pages

Abstract: In social-learning settings where individuals receive private signals andobserve network neighbors actions, the network structure often obstructsinformation aggregation. We consider sequential social learning with rationalagents and Gaussian signals and ask how the efficiency of signal aggregationchanges with the network. Rational actions in our model admit a signal-countinginterpretation of accuracy, which lets us compare the aggregative efficiency ofsocial learning across networks. Learning is very inefficient in a class ofnetworks where agents move in generations and observe the previous generation.Generations after the first contribute very little additional information, evenwhen generations are arbitrarily large.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×