Preliminary Results on Social Learning with Partial Observations

Ilan Lobel; Daron Acemoglu; Munther Dahleh; Asuman Ozdaglar

1st International ICST Workshop on interdisciplinary systems approach in performance evaluation and design of computer and comunication system

Research Article

Preliminary Results on Social Learning with Partial Observations

Download519 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.4108/inter-perf.2007.2074,
    author={Ilan Lobel and Daron Acemoglu and Munther Dahleh and Asuman Ozdaglar},
    title={Preliminary Results on Social Learning with Partial Observations},
    proceedings={1st International ICST Workshop on interdisciplinary systems approach in performance evaluation and design of computer and comunication system},
    proceedings_a={INTER-PERF},
    year={2010},
    month={5},
    keywords={},
    doi={10.4108/inter-perf.2007.2074}
}

Ilan Lobel
Daron Acemoglu
Munther Dahleh
Asuman Ozdaglar
Year: 2010
Preliminary Results on Social Learning with Partial Observations
INTER-PERF
ICST
DOI: 10.4108/inter-perf.2007.2074

Ilan Lobel¹^,*, Daron Acemoglu²^,*, Munther Dahleh³^,*, Asuman Ozdaglar³^,*

1: Operations Research Center at the Massachusetts Institute of Technology, Cambridge, MA.
2: Department of Economics at the Massachusetts Institute of Technology, Cambridge, MA.
3: Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology, Cambridge, MA.

*Contact email: lobel@mit.edu, daron@mit.edu, dahleh@mit.edu, asuman@mit.edu

Abstract

We study a model of social learning with partial observations from the past. Each individual receives a private signal about the correct action he should take and also observes the action of his immediate neighbor. We show that in this model the behavior of asymptotic learning is characterized in terms of two threshold values that evolve deterministically. Individual actions are fully determined by the value of their signal relative to these two thresholds. We prove that asymptotic learning from an ex ante viewpoint applies if and only if individual beliefs are unbounded. We also show that symmetry between the states implies that the minimum possible amount of asymptotic learning occurs.

Published: 2010-05-16
Modified: 2011-09-15

: http://dx.doi.org/10.4108/inter-perf.2007.2074