Costly Information, Imperfect Learning, and Information Aggregation

Abstract

Underlying the emergence of the incomplete-information literature is the insight that real-world investors do not know assets’ expected returns. Modelling the information acquisition process explicitly, this literature aims at making models of dynamic portfolio choice and asset pricing more realistic and more accurate. However, all of the papers reviewed in Chap. 1, as well as the previous chapters of this study, assume — at least implicitly — that parameter estimation can be performed costlessly and therefore perfectly, and use optimal filtering theory to model the economic agents’ inference process.