Partial Identification of Probability Distributions

By Charles F. Manski. 2003

760 citations recorded by the source corpus.

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

The Econometrics of Randomized Experiments (2016, background): “…treatment in settings with non-compliance an alternative to making additional assumptions, is to focus on getting ranges of values for the estimand that are consistent with the data in a bounds or partial identification approach in a line of research associated with Manski (1990, 1996, 2003, 2013).”
Bounds on causal effects in three‐arm trials with non‐compliance (2006, background): “…are only partially identified under certain assumptions, finding the upper and lower bounds of the identification region of the parameter establishes a domain of consensus among researchers who may hold disparate beliefs about what other assumptions are appropriate (Joffe, 2001; Manski, 2003).”
The Economics of Reproducibility in Preclinical Research (2015, background): “Using a highly conservative probability bounds approach [17], we estimate that the cumulative rate of preclinical irreproducibility lies between 18% (the maximum of the low estimates, assuming maximum overlap between categories), and 88.”
Set Identified Linear Models (2011, background): “There are many other examples of missing information generating incomplete identification (see Manski, 2003).”
Identification of treatment response with social interactions (2012, methodology): “in Manski (1990, 1997, 2003) , Manski and Pepper (2000), and elsewhere.”
Confidence Intervals for Partially Identified Parameters (2003, background): “In the last decade a growing body of research has studied inference on partially identified parameters (e.g., Manski, 1990, 2003 ).”
Instrumental Variables: An Econometrician's Perspective (2014, positive): “The initial results in this literature demonstrated the difficulties in establishing point identification (Heckman, 1990; Manski, 1990), leading to the bounds approach developed by Manski (1995, 2003).”
Sign Restrictions, Structural Vector Autoregressions, and Useful Prior Information ∗ (2015, background): “There is a huge literature that considers econometric inference under set identification using a frequentist approach; see for example the reviews in Manski (2003) and Tamer (2010).”
Inference on Directionally Differentiable Functions (2014, background): “37 63 v1 [ m at h. ST ] 1 4 A pr 2”
PARTIAL IDENTIFICATION OF THE DISTRIBUTION OF TREATMENT EFFECTS AND ITS CONFIDENCE SETS (2009, background): “Bounds on D-Parameters The sharp bounds on the treatment effect distribution implies bounds on the class of ‘‘D-parameters’’ introduced in Manski (1997a), see also Manski (2003).”
Bayesian inference in a class of partially identified models (2016, background): “In particular, “simple interval identiﬁed parameters” concerns d µ = 2, and arises in the context of missing data and general “selection problems” (e.g., Manski (2003)) and best response functions in games (e.g., Kline and Tamer (2012)).”
Minimax regret treatment choice with covariates or with limited validity of experiments (2012, background): “For the purpose of treatment choice, what is important about all of these problems is the precise wedge they drive between observable treatment outcomes and treatment outcomes that 11See Manski (2003) for a survey.”
Instrumental variable methods for recovering continuous linear functionals (2011, background): “i=1 ( yiθ̂(zi)− E[Y θ̂(Z)] ) + √ nE[Y (θ̂(Z)− θ0(Z))] (15)”
Moving the Goalposts: Addressing Limited Overlap in Estimation of Average Treatment Effects by Changing the Estimand (2006, neutral): “In cases where even large data sets would not permit point identification of the estimand or interest regions of the parameter space consistent with the model may be reported in a bounds analysis of the type developed by Manski (1990, 2003) .”