Skip to page navigation menu Skip entire header
Brown University
Skip 13 subheader links

Essays on the Use of Information in Estimation and Effort Motivation

Description

Abstract:
This dissertation studies the use of information in Econometrics and as a tool for effort motivation. Chapter 1 considers the question of whether there exist population versions of the finite sample implied probabilities that emerge in the optimization problem of the Empirical Cressie-Read estimators. The main result shows that such an existence necessitates strong assumptions which vary depending on which estimator is used. In particular, if a boundedness assumption is not made for the moment function when using Empirical Likelihood, the pseudo-true value does not exist rendering robustness to misspecification impossible. Chapter 2 combines this result with a result by Newey and Smith who show that Empirical Likelihood is higher order efficient to say that no member of the Empirical Cressie-Read class of estimators can be both robust to misspecification with reasonable assumptions as well as higher order efficient. As a solution to this problem a new class of estimators dubbed Higher Order Efficient Bounded Tilting Estimators is introduced which require no assumptions on the data generating process while also being higher order efficient. Chapter 3 moves away from Econometric Theory to examine the problem of a principal attempting to incentivize an agent to provide effort using information instead of monetary compensation. The theoretically optimal disclosure policy for the principal involves the use of both a delay and obfuscation of the information. We design and execute an experiment where we find significant deviations from the theoretically optimal responses, which results in an information disclosure policy which delays information but avoids obfuscation to perform the best. We also find that reciprocity does not play a role in the decisions of the agents, and that the results could better be explained by a heterogeneity in risk attitudes of the subjects.
Notes:
Thesis (Ph. D.)--Brown University, 2020

Citation

Wahlstrom, Oscar Jonas, "Essays on the Use of Information in Estimation and Effort Motivation" (2020). Economics Theses and Dissertations. Brown Digital Repository. Brown University Library. https://repository.library.brown.edu/studio/item/bdr:5vqkx7ck/

Relations

Collection: