Monotonicity conditions and inequality imputation for sample-selection and non-response problems

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Under a sample selection or non-response problem, where a response variable y is observed only when a condition δ = 1 is met, the identified mean E(y|δ = 1) is not equal to the desired mean E(y). But the monotonicity condition E(y|δ = 1) ≤ E(y|δ = 0) yields an informative bound E(y|δ = 1) ≤ E(y), which is enough for certain inferences. For example, in a majority voting with δ being the vote-turnout, it is enough to know if E(y) > 0.5 or not, for which E(y|δ = 1) > 0.5 is sufficient under the monotonicity. The main question is then whether the monotonicity condition is testable, and if not, when it is plausible. Answering to these queries, when there is a 'proxy' variable z related to y but fully observed, we provide a test for the monotonicity; when z is not available, we provide primitive conditions and plausible models for the monotonicity. Going further, when both y and z are binary, bivariate monotonicities of the type P(y, z|δ = 1) ≤ P(y, z|δ = 0) are considered, which can lead to sharper bounds for P(y). As an empirical example, a data set on the 1996 U.S. presidential election is analyzed to see if the Republican candidate could have won had everybody voted, i.e., to see if P(y) > 0.5, where y = 1 is voting for the Republican candidate.

Original languageEnglish
Pages (from-to)175-194
Number of pages20
JournalEconometric Reviews
Volume24
Issue number2
DOIs
Publication statusPublished - 2005

Keywords

  • Imputation
  • Monotonicity
  • Non-response
  • Orthant dependence
  • Sample selection

ASJC Scopus subject areas

  • Economics and Econometrics

Fingerprint Dive into the research topics of 'Monotonicity conditions and inequality imputation for sample-selection and non-response problems'. Together they form a unique fingerprint.

  • Cite this