Full Citation

Title: Ecological Regression with Partial Identification

Citation Type: Miscellaneous

Publication Year: 2018

ISBN:

ISSN:

DOI:

NSFID:

PMCID:

PMID:

Abstract: We study a partially identified linear contextual effect model for ecological inference, and describe how to perform inference on the district level parameter averaging over many precincts, in the presence of the unidentified parameter of the contextual effect. We derive various bounds for this unidentified parameter of contextual effect, from the tightest possible, to ones that are more convenient to use. This may be regarded as a first attempt to limit the scope of non-identifiability in linear contextual effect models. As an application, the linear contextual model implies a “regression bound” for the district level parameter of interest, which can intersect the model-free bound by Duncan and Davis (1953) and achieve a shorter length.

Url: https://gking.harvard.edu/files/gking/files/ei-wkst.pdf

User Submitted?: No

Authors: Jiang, Wenxin; King, Gary; Schmaltz, Allen; Tanner, Martin, A

Publisher: Harvard University

Data Collections: IPUMS USA

Topics: Methodology and Data Collection, Other

Countries: