Shape‐restricted statistical inference for non‐ignorable missing data under a general additive model

Although ubiquitous in many areas, missing data problems become challenging when the missingness of the outcome depends on itself, which means the data are non‐ignorable missing. To alleviate the risk of model misspecification and balance interpretation and efficiency, we model the non‐missingness probability by a logistic model with a general additive covariate effect under shape restrictions. Each additive component is assumed to satisfy certain shape restrictions, such as monotone increasing/decreasing, convexity/concavity, or a combination of these. We develop a shape‐restricted and tuning‐parameter‐free estimator for the population outcome mean with the help of an instrument variable. We systematically establish the consistency, convergence rates, and asymptotic normalities of the proposed estimators. Our numerical results indicate that the proposed shape‐restricted estimator has comparable performance to competing estimators with parametric models when the parametric models are correct, and outperforms them when the parametric models are misspecified. Finally, our method is applied to two real datasets providing more interpretable results than its competitors.

• Publication year

  2026

• Venue

  Scandinavian Journal of Statistics

• Publication date

  2026-01-06

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1111/sjos.70051
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Robust estimation under a semiparametric propensity model for nonignorable missing data

  2025cited by this paper
• Verifiable identification condition for nonignorable nonresponse data with categorical instrumental variables

  2023cited by this paper
• Instability of inverse probability weighting methods and a remedy for nonignorable missing data

  2023influential reference
• Statistical inference with semiparametric nonignorable nonresponse models

  2023cited by this paper
• Score test for missing at random or not under logistic missingness models

  2022cited by this paper
• Non-parametric inference about mean functionals of non-ignorable non-response data without identifying the joint distribution.

  2021cited by this paper
• Efficient estimation in a partially specified nonignorable propensity score model

  2021cited by this paper
• Nonignorable item nonresponse in panel data

  2020cited by this paper
• Multidimensional Shape Constraints

  2020cited by this paper
• Full-semiparametric-likelihood-based inference for non-ignorable missing data

  2019influential reference
• Statistical inference for nonignorable missing-data problems: a selective review

  2018cited by this paper
• A Propensity-score-adjustment Method for Nonignorable Nonresponse

  2016cited by this paper
• Semiparametric optimal estimation with nonignorable nonresponse data

  2016cited by this paper
• Semiparametric inverse propensity weighting for nonignorable missing data

  2016cited by this paper
• Semiparametric Estimation with Data Missing Not at Random Using an Instrumental Variable.

  2016cited by this paper
• Semiparametric Pseudo-Likelihoods in Generalized Linear Models With Nonignorable Missing Data

  2015cited by this paper
• On varieties of doubly robust estimators under missingness not at random with a shadow variable

  2015cited by this paper
• Identifiability of Normal and Normal Mixture Models with Nonignorable Missing Data

  2015cited by this paper
• Empirical Likelihood for Estimating Equations with Nonignorably Missing Data.

  2014cited by this paper
• An Instrumental Variable Approach for Identification and Estimation with Nonignorable Nonresponse

  2014influential reference
• Generalized additive and index models with shape constraints

  2014influential reference
• Nonparametric Estimation under Shape Constraints: Estimators, Algorithms and Asymptotics

  2014cited by this paper
• Statistical Methods for Handling Incomplete Data

  2013cited by this paper
• A unifying view on dataset shift in classification

  2012cited by this paper
• A Semiparametric Estimation of Mean Functionals With Nonignorable Missing Data

  2011influential reference
• Using Calibration Weighting to Adjust for Nonignorable Unit Nonresponse

  2010cited by this paper
• Using Calibration Weighting to Adjust for Nonresponse Under a Plausible Model (with full appendices)

  2007cited by this paper
• Measuring and testing dependence by correlation of distances

  2007influential reference
• Generalized Additive Models

  2006cited by this paper
• Mathematical Statistics

  2006cited by this paper
• Stable and Efficient Multiple Smoothing Parameter Estimation for Generalized Additive Models

  2004cited by this paper
• Analysis of multivariate missing data with nonignorable nonresponse

  2003cited by this paper
• Nonparametric option pricing under shape restrictions

  2003cited by this paper
• Estimation With Survey Data Under Nonignorable Nonresponse or Informative Sampling

  2002cited by this paper
• Sample Selection Bias Correction for Missing Response Observations

  2000cited by this paper
• Observed information in semi-parametric models

  1999influential reference
• The Reduced Monotonic Regression Method

  1997cited by this paper
• Toward a curse of dimensionality appropriate (CODA) asymptotic theory for semi-parametric models.

  1997cited by this paper
• Applied Logistic Regression, Second Edition

  1989influential reference
• Regression Analysis for Categorical Variables with Outcome Subject to Nonignorable Nonresponse

  1988cited by this paper
• Imputation of Missing Values When the Probability of Response Depends on the Variable Being Imputed

  1982cited by this paper
• INFERENCE AND MISSING DATA

  1975cited by this paper

• No citing papers are available for this paper.