Nonparametric Analysis of Randomized Experiments with Missing Covariate and Outcome Data

By Joel L. Horowitz, Charles F. Manski. 2000

476 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/948ad7ff478d48500350278e93c6559c06f7cc2a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

More on monotone instrumental variables (2009, unknown): “Moreover, the methods for forming asymptotically valid confidence sets for partially identified parameters developed by Horowitz and Manski (2000), Imbens and Manski (2004), Chernozhukov, Hong, and Tamer (2007), Beresteanu and Molinari (2008), Rosen (2006), and others appear not to apply.”
The Underutilized Potential of Teacher-to-Parent Communication: Evidence from a Field Experiment (2015, methodology): “Lee bounds also provide more narrow ranges than the worst-case imputation procedure developed by Horowitz and Manski (2000).”
Bounds on causal effects in three‐arm trials with non‐compliance (2006, background): “Ln −zÅnα, Un +zÅnα/ (Horowitz and Manski, 2000), which has asymptotic coverage probability 1−α (Bickel and Freedman, 1981).”
Keeping It Simple: Financial Literacy and Rules of Thumb (2014, methodology): “…on the treatment effects for the rule-of-thumb training on the business practices index, revenue index, and any reporting errors indicator using a range of assumptions for the pattern of attrition following an approach based on Horowitz and Manski (2000), Lee (2002), and Kling and Liebman (2004).”
Teaching Statistical Inference for Causal Effects in Experiments and Observational Studies (2004, background): “Some relevant references for this approach include Manski et al. (1992), Manski and Nagin (1998), and Horowitz and Manski (2000).”
Teaching Entrepreneurship: Impact of Business Training on Microfinance Clients and Institutions (2011, methodology): “Given our nonresponse rate, we analyze the implications of different plausible assumptions on our estimated treatment effects, along the lines of the nonparametric approaches followed in Horowitz and Manski (2000) and Lee (2002).”
Bayesian Analysis in Moment Inequality Models (2010, background): “It follows that B(θ2, R4 ) ⊂ B(θ1,R) ⊂ {θ ∈ :g(θ) ≤ infθ∈ g( )−ε} and θ2 ∈ A.”
Returns to microcredit, cash grants and training for male and female microentrepreneurs in Uganda (2018, unknown): “In columns (5) to (10) I present a bounding exercise similar to that conducted by Karlan and Valdivia (2011), who use a range of assumptions for bounding originally from Horowitz and Manski (2000), Lee (2002) and Kling and Liebman (2007). New lower-effect bounds are created by imputing the outcomes ”
Partial Identification in Monotone Binary Models: Discrete Regressors and Interval Data (2008, result): “As for references, this paper belongs to the small, but growing literature on partial identication as pionneered by Manski (2003, and references therein). Our results on bounds on parameters in binary regressions can be seen as generalizations of bounds on averages, derived in the paper by Green, J”
More on Confidence Intervals for Partially Identified Parameters (2008, background): “Pr Ã −l + √ N∆∗ b σl + q 1− bρ2z2 ≤ bρz1, z1 ≤ cu b σu ! ≥ 1− α, (9)”
Social Experiments in the Labor Market (2016, unknown): “One bounding approach is proposed by Horowitz and Manski (2000). This strategy asks how much the estimated treatment effect would be inflated if all missing treatment observations were assumed to have the highest possible outcomes and all missing control observations the lowest; then it asks how muc”