Statistical Analysis With Missing Data

By Subir Ghosh. 1988

How researchers see this paper

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

Citations in the corpus

Recommendations as Treatments: Debiasing Learning and Evaluation (2016, methodology): “Maximum likelihood estimates for P ( Y = r | O = 1) and P ( O = 1) can be obtained by counting observed ratings in MNAR data.”
Causal Transfer in Machine Learning (2015, methodology): “Chapter 11 in Little and Rubin [1986] provides the update equations for optimizing Equation (8) using EM.”
The Selective Labels Problem: Evaluating Algorithmic Predictions in the Presence of Unobservables (2017, methodology): “Prior research also categorized missing data problems based on the reasons behind the missingness [26, 28].”
Neural networks for pattern recognition (1995, background): “A common problem is that some of the input values may be missing from the data set for some of the pattern vectors (Little and Rubin, 1987; Little, 1992).”
Making Sense of Issues Through Media Frames: Understanding the Kosovo Crisis (2006, methodology): “These data transcription errors appear to be random; we discarded these participants from the analysis and treated them as missing at random (Little and Rubin 1987). noitidno C acire m A o T ksi R noitidno C lortno C noitidno C sisir C nairatina mu H C lu st er 1 : N A T O , U N , P ea ce ke ep er…”
Efficient Estimation of Average Treatment Effects Using the Estimated Propensity Score 1 (2002, background): “Robins, Mark and Newey (1992), Robins and Rotnitzky (1995), Robins, Rotnitzky and Zhao (1995), and Rotnitzky and Robins (1995) study the related problem of inference for parameters in regression models where some data are Missing At Random (MAR, Rubin, 1976; Little and Rubin, 1987).”
Impact of imputation of missing values on classification error for discrete data (2008, methodology): “…based methods work based on parameter estimation in the presence of missing data, i.e., the data are described based on the models and their parameters are estimated by maximum likelihood or maximum a posteriori procedures that use variants of the expectation maximization algorithm [3] [6].”
Labor Market Competition and Individual Preferences Over Immigration Policy (2001, background): “The most general and extensively researched approach is "multiple imputation" (King, et al (1998a), Schafer (1997), Little and Rubin (1987), Rubin (1987)).”
Principles of Data Mining (2001, background): “Luenberger (1984) and Nering and Tucker (1993) discuss linear programming and related constrained optimization techniques in detail. Mangasarian (1997) describes the application of constrained optimization techniques to a variety problems in data mining, including feature selection, clustering, and ”
Random Effects Models for Longitudinal Data (2010, methodology): “When referring to the missing-value, or non-response, process we will use the terminology of Little and Rubin (2002). A non-response process is said to be missing completely at random (MCAR), if the missingness is independent of both unob-”
Good research practices for cost-effectiveness analysis alongside clinical trials: the ISPOR RCT-CEA Task Force report. (2005, background): “Missing data may bear no relation to observed or unobserved factors in the population (missing completely at random), may have a relationship to observed variables (missing at random), or may be related to unobserved factors (not missing at random) [ 66 ].”