Nonparametric Analysis of Randomized Experiments with Missing Covariate and Outcome Data

Horowitz, Joel L.; Manski, Charles F.

doi:10.2307/2669526

Cited by 190 publications

(274 citation statements)

References 16 publications

Supporting

Mentioning

273

Contrasting

Order By: Relevance

“…Bickel and Friedman (1981) proved that the bootstrap can be used to construct confidence intervals for two unknown parameters simultaneously. Horowitz and Manski (2000) use the bootstrap to put bounds on the treatment effect for missing-value data, where either baseline covariates and/or outcomes are missing for some subjects. The same method was used to provide a joint confidence interval for a pair of lower and upper cluster bounds on the parameter π 01.…”

Section: Property Assessmentmentioning

confidence: 99%

Properties Of Bound Estimators On Treatment Effect Heterogeneity For Binary Outcomes

Mascha

Albert²

2006

J. Mod. App. Stat. Meth.

View full text Add to dashboard Cite

Variability in individual causal effects, treatment effect heterogeneity (TEH), is important to the interpretation of clinical trial results, regardless of the marginal treatment effect. Unfortunately, it is usually ignored. In the setting of two-arm randomized studies with binary outcomes, there are estimators for bounds on the probability of control success and treatment failure for an individual, or the treatment risk. Here, those bounds were refined and the sampling properties were assessed using simulations of correlated multinomial data via the Dirichlet multinomial. Results indicated low bias and mean squared error. Moderate to high intraclass correlation (ICC) and large numbers of clusters allow narrower confidence interval widths for the treatment risk.

show abstract

Section: Property Assessmentmentioning

confidence: 99%

Properties Of Bound Estimators On Treatment Effect Heterogeneity For Binary Outcomes

Mascha

Albert²

2006

J. Mod. App. Stat. Meth.

View full text Add to dashboard Cite

show abstract

“…The second result shows that, for interval identi…ed , asymptotic equivalence holds between the posterior lower probability inference and the frequentist inference for the identi…ed set considered in Horowitz and Manski (2000), Chernozhukov, Hong, and Tamer (2007), and Romano and Shaikh (2010). Note that the posterior lower credible region C cannot be interpreted as the frequentist con…dence interval for the parameter of interest considered in Imbens and Manski (2004).…”

Section: Condition 52 (I)mentioning

confidence: 92%

Inference and decision for set identified parameters using posterior lower and upper probabilities

Kitagawa

2011

View full text Add to dashboard Cite

This paper develops inference and statistical decision for set-identi…ed parameters from the robust Bayes perspective. When a model is set-identi…ed, prior knowledge for model parameters is decomposed into two parts: the one that can be updated by data (revisable prior knowledge) and the one that never be updated (unrevisable prior knowledge.) We introduce a class of prior distributions that shares a single prior distribution for the revisable, but allows for arbitrary prior distributions for the unrevisable. A posterior inference procedure proposed in this paper operates on the resulting class of posteriors by focusing on the posterior lower and upper probabilities. We analyze point estimation of the set-identi…ed parameters with applying the gamma-minimax criterion. We propose a robusti…ed posterior credible region for the set-identi…ed parameters by focusing on a contour set of the posterior lower probability. Our framework o¤ers a procedure to eliminate set-identi…ed nuisance parameters, and yields inference for the marginalized identi…ed set. For an interval identi…ed parameter case, we establish asymptotic equivalence of the lower probability inference to frequentist inference for the identi…ed set.Keywords: Partial Identi…cation, Bayesian Robustness, Belief Function, Imprecise Probability, Gamma-minimax, Random Set.JEL Classi…cation: C12, C15, C21.

show abstract

“…150 (3) 4 Horowitz and Manski (2000) derived the analytical solution for the optimization problem in the case of binary outcomes. Our problem here is however more complex.…”

Section: Estimation Of Boundsmentioning

confidence: 99%

Obtaining and Predicting the Bounds of Realized Correlations

Grossmass

2014

Swiss J Economics Statistics

View full text Add to dashboard Cite

Summary This paper argues that the inherent data problems make precise point identification of realized correlation difficult but identification bounds in the spirit of Manski (1995) can be derived. These identification bounds allow for a more robust approach to inference especially when the realized correlation is used for estimating other risk measures. We forecast the identification bounds using the HAR model of Corsi (2003) using data during the year of onset of the credit crisis and find that the bounds provide good predictive coverage of the realized correlation for both 1- and 10-step forecasts even in volatile periods.

show abstract

Nonparametric Analysis of Randomized Experiments with Missing Covariate and Outcome Data

Cited by 190 publications

References 16 publications

Properties Of Bound Estimators On Treatment Effect Heterogeneity For Binary Outcomes

Properties Of Bound Estimators On Treatment Effect Heterogeneity For Binary Outcomes

Inference and decision for set identified parameters using posterior lower and upper probabilities

Obtaining and Predicting the Bounds of Realized Correlations

Contact Info

Product

Resources

About