Confidence Intervals for Partially Identified Parameters

2004/10/08 by Guido W. Imbens, Charles F. Manski · 30 citations
Mathematics · #Statistical Methods and Inference #Statistical Methods in Clinical Trials #Statistical Methods and Bayesian Inference

paper · doi:10.1111/j.1468-0262.2004.00555.x

Abstract

this paper, we study the use of these intervals as CIs for the partially identified parameter f(P,#). Our most basic finding is Lemma 2.1: Lemma 2.1 Let CN0 0, CN1 0, # #, and P #P

Cited by

• Recent Developments in the Econometrics of Program Evaluation
• Generalizing Trimming Bounds for Endogenously Missing Outcome Data Using Random Forests
• Plausibly Exogenous
• Identifying the Effects of WIC on Food Insecurity Among Infants and Children
• Noise-induced randomization in regression discontinuity designs
• Nonparametric Tests of Panel Conditioning and Attrition Bias in Panel Surveys
• Tightening belts to buy a home: Consumption responses to rising housing prices in urban China
• Compositional difference-in-differences
• Partial Identification of Causal Effects that Vary by Setting
• PLRD: Partially Linear Regression Discontinuity Inference
• Revision Risk in Real-Time Macroeconomic Forecasting
• Party On: The Labor Market Returns to Social Networks in Adolescence
• Inference in partially identified moment models via regularized optimal transport
• Microfoundations and the Causal Interpretation of Price-Exposure Designs
• Difference-in-Differences with Interval Data
• A Sensitivity Approach to Causal Inference Under Limited Overlap
• Generalized method of moments with partially missing data
• The use of uncertainty to choose matching variables in statistical matching
• Should data ever be thrown away? Pooling interval-censored data sets with different precision
• Semi-parametric Bayesian Partially Identified Models based on Support\n Function
• Spillover Effects in Cluster Randomized Trials with Noncompliance
• Don't Cross the Line: Bounding the Causal Effect of Hypergamy Violation on Domestic Violence in India
• Toward Credible Patient-centered Meta-analysis
• Non-overlap Average Treatment Effect Bounds
• More than Just Friends? School Peers and Adult Interracial Relationships
• Treatment Effects of Multi-Valued Treatments in Hyper-Rectangle Model
• Non-Existent Outcomes in Research on Inequality: A Causal Approach
• Measuring economic mobility in India using noisy data: a partial identification approach
• Partial identification of mean achievement in ILSA studies with multi-stage stratified sample designs and student non-participation
• A Non-Parametric Sensitivity Analysis for Bounding Bias in Hybrid Control Trials

Related

• On the power properties of inference for parameters with interval identified sets
• A Simple, Short, but Never-Empty Confidence Interval for Partially\n Identified Parameters
• Optimal confidence intervals for bounded parameters (a correct alternative to the recipe of Feldman and Cousins)
• Optimal upper bounds for non-negative parameters
• Confidence belts on bounded parameters