CaltechAUTHORS
  A Caltech Library Service

Optimal Data Acquisition for Statistical Estimation

Chen, Yiling and Immorlica, Nicole and Lucier, Brendan and Syrgkanis, Vasilis and Ziani, Juba (2018) Optimal Data Acquisition for Statistical Estimation. In: Proceedings of the 2018 ACM Conference on Economics and Computation. Association for Computing Machinery , New York, NY, pp. 27-44. ISBN 978-1-4503-5829-3. http://resolver.caltech.edu/CaltechAUTHORS:20180828-141500283

[img] PDF - Published Version
See Usage Policy.

1344Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20180828-141500283

Abstract

We consider a data analyst's problem of purchasing data from strategic agents to compute an unbiased estimate of a statistic of interest. Agents incur private costs to reveal their data and the costs can be arbitrarily correlated with their data. Once revealed, data are verifiable. This paper focuses on linear unbiased estimators. We design an individually rational and incentive compatible mechanism that optimizes the worst-case mean-squared error of the estimation, where the worst-case is over the unknown correlation between costs and data, subject to a budget constraint in expectation. We characterize the form of the optimal mechanism in closed-form. We further extend our results to acquiring data for estimating a parameter in regression analysis, where private costs can correlate with the values of the dependent variable but not with the values of the independent variables.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1145/3219166.3219195DOIArticle
Additional Information:© 2018 Copyright held by the owner/author(s). Publication rights licensed to ACM. Yiling Chen was partially supported by NSF grant CCF-1718549. Juba Ziani was supported by NSF grants CNS-1331343 and CNS-1518941, and the US-Israel Binational Science Foundation grant 2012348. Part of the work was done while Yiling Chen and Juba Ziani were at Microsoft Research New England.
Funders:
Funding AgencyGrant Number
NSFCCF-1718549
NSFCNS-1331343
NSFCNS-1518941
Binational Science Foundation (USA-Israel)2012348
Microsoft ResearchUNSPECIFIED
Subject Keywords:buying data; budget-feasible mechanism design; statistical estimation; optimization
Record Number:CaltechAUTHORS:20180828-141500283
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20180828-141500283
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:89258
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:28 Aug 2018 22:20
Last Modified:28 Aug 2018 22:20

Repository Staff Only: item control page