TY - JOUR
T1 - The value of information - An axiomatic approach
AU - Gilboa, Itzhak
AU - Lehrer, Ehud
N1 - Funding Information:
*We are grateful to Ehud Kalai, Morton Kamien, Roger Myerson, Dov Samet, two anonymous referees, and especially Dov Monderer for helpful discussions. The first author gratefully acknowledges NSF Grant No. IRI-8814672, and the use of o!Xce facilities at the Cowles Foundation, Yale University.
PY - 1991
Y1 - 1991
N2 - The following question is addressed: Given a real-valued function on the partitions of a measure space, what are necessary and sufficient conditions in order that it be the value of information function for a Bayesian decision maker? A characterization is provided, and the analysis reveals a close relationship to cooperative game theory. The tools developed for characterization can also be used to derive somewhat surprising results regarding the properties of information functions.
AB - The following question is addressed: Given a real-valued function on the partitions of a measure space, what are necessary and sufficient conditions in order that it be the value of information function for a Bayesian decision maker? A characterization is provided, and the analysis reveals a close relationship to cooperative game theory. The tools developed for characterization can also be used to derive somewhat surprising results regarding the properties of information functions.
UR - http://www.scopus.com/inward/record.url?scp=0347336649&partnerID=8YFLogxK
U2 - 10.1016/0304-4068(91)90002-B
DO - 10.1016/0304-4068(91)90002-B
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:0347336649
SN - 0304-4068
VL - 20
SP - 443
EP - 459
JO - Journal of Mathematical Economics
JF - Journal of Mathematical Economics
IS - 5
ER -