Robust Likelihood-based Analysis of Multivariate Data with Missing Values

By Roderick J. A. Little, H. An. 2003

172 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/8d57ab403b0e4fc8ac91e6cc0555f2531d09054b

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

On the Construction of Imputation Classes in Surveys (2007, methodology): “Finally, there are non-parametric or semi-parametric methods that have recently been proposed and that also address the robustness issue (e.g. Scharfstein & Irizarry, 2003 or Little & An, 2004).”
Inference for the Population Total from Probability-Proportional-to-Size Samples Based on Predictions from a Penalized Spline Nonparametric Model (2003, unknown): “His approach is discussed in the context of missing-data adjustments in Little and An (2004).”
A Primer on Propensity Score Analysis (2010, methodology): “Hence, regression approaches that are less sensitive to functional form assumptions were suggested.(22,29) These approaches rely on cubic spline regression or dummies based on a categorized propensity score.”
Bayesian Regression Tree Models for Causal Inference: Regularization, Confounding, and Heterogeneous Effects (with Discussion) (2020, unknown): “Little and An, 2004; Zhou et al., 2019).”
Analysis of Incomplete Data Using Inverse Probability Weighting and Doubly Robust Estimators (2010, methodology): “…estimator, without sacrificing too much precision, it has been suggested – with some good results – that one should reduce multiple background characteristics Xi into a single characteristic πi which may then be modeled more flexibly (David, Little, Samuhel & Triest, 1983; Little & An, 2004).”
Comment: Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data. (2008, methodology): “Little and An (2004) incorporated a smoothing spline hased on η̂i and demonstrated that the resulting regression estimate of μ is DR in the following sense: If the y-model correctly describes E(yi | xi) before the propensity-related terms are added, these additional terms (or any other functions of ”