Dilation bootstrap

Alfred Galichon, Marc Albert Henry

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

We propose a methodology for combining several sources of model and data incompleteness and partial identification, which we call Composition Theorem. We apply this methodology to the construction of confidence regions with partially identified models of general form. The region is obtained by inverting a test of internal consistency of the econometric structure. We develop a dilation bootstrap methodology to deal with sampling uncertainty without reference to the hypothesized economic structure. It requires bootstrapping the quantile process for univariate data and a novel generalization of the latter to higher dimensions. Once the dilation is chosen to control the confidence level, the unknown true distribution of the observed data can be replaced by the known empirical distribution and confidence regions can then be obtained as in Galichon and Henry (2011) and Beresteanu et al. (2011).

Original languageEnglish (US)
Pages (from-to)109-115
Number of pages7
JournalJournal of Econometrics
Volume177
Issue number1
DOIs
StatePublished - Jan 1 2013

Fingerprint

Dilation
Bootstrap
Confidence Region
Methodology
Partial Identification
Quantile Process
Internal Consistency
Identification (control systems)
Henry
Incompleteness
Empirical Distribution
Bootstrapping
Confidence Level
Sampling
Econometrics
Economics
Higher Dimensions
Univariate
Chemical analysis
Uncertainty

All Science Journal Classification (ASJC) codes

  • Economics and Econometrics
  • Applied Mathematics
  • History and Philosophy of Science

Cite this

Galichon, Alfred ; Henry, Marc Albert. / Dilation bootstrap. In: Journal of Econometrics. 2013 ; Vol. 177, No. 1. pp. 109-115.
@article{0a6b0f1982b6471d92069cf14903fd8d,
title = "Dilation bootstrap",
abstract = "We propose a methodology for combining several sources of model and data incompleteness and partial identification, which we call Composition Theorem. We apply this methodology to the construction of confidence regions with partially identified models of general form. The region is obtained by inverting a test of internal consistency of the econometric structure. We develop a dilation bootstrap methodology to deal with sampling uncertainty without reference to the hypothesized economic structure. It requires bootstrapping the quantile process for univariate data and a novel generalization of the latter to higher dimensions. Once the dilation is chosen to control the confidence level, the unknown true distribution of the observed data can be replaced by the known empirical distribution and confidence regions can then be obtained as in Galichon and Henry (2011) and Beresteanu et al. (2011).",
author = "Alfred Galichon and Henry, {Marc Albert}",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.jeconom.2013.07.001",
language = "English (US)",
volume = "177",
pages = "109--115",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "1",

}

Dilation bootstrap. / Galichon, Alfred; Henry, Marc Albert.

In: Journal of Econometrics, Vol. 177, No. 1, 01.01.2013, p. 109-115.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dilation bootstrap

AU - Galichon, Alfred

AU - Henry, Marc Albert

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We propose a methodology for combining several sources of model and data incompleteness and partial identification, which we call Composition Theorem. We apply this methodology to the construction of confidence regions with partially identified models of general form. The region is obtained by inverting a test of internal consistency of the econometric structure. We develop a dilation bootstrap methodology to deal with sampling uncertainty without reference to the hypothesized economic structure. It requires bootstrapping the quantile process for univariate data and a novel generalization of the latter to higher dimensions. Once the dilation is chosen to control the confidence level, the unknown true distribution of the observed data can be replaced by the known empirical distribution and confidence regions can then be obtained as in Galichon and Henry (2011) and Beresteanu et al. (2011).

AB - We propose a methodology for combining several sources of model and data incompleteness and partial identification, which we call Composition Theorem. We apply this methodology to the construction of confidence regions with partially identified models of general form. The region is obtained by inverting a test of internal consistency of the econometric structure. We develop a dilation bootstrap methodology to deal with sampling uncertainty without reference to the hypothesized economic structure. It requires bootstrapping the quantile process for univariate data and a novel generalization of the latter to higher dimensions. Once the dilation is chosen to control the confidence level, the unknown true distribution of the observed data can be replaced by the known empirical distribution and confidence regions can then be obtained as in Galichon and Henry (2011) and Beresteanu et al. (2011).

UR - http://www.scopus.com/inward/record.url?scp=84883446946&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883446946&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2013.07.001

DO - 10.1016/j.jeconom.2013.07.001

M3 - Article

AN - SCOPUS:84883446946

VL - 177

SP - 109

EP - 115

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 1

ER -