Spline Pattern-Mixture Models for Missing Data

Yang, Ye; Little, Roderick J.A.

doi:10.6339/21-JDS1008

Journal of Data Science

Spline Pattern-Mixture Models for Missing Data

Volume 19, Issue 1 (2021), pp. 75–95

Ye Yang Roderick J.A. Little

https://doi.org/10.6339/21-JDS1008

Pub. online: 10 February 2021 Type: Statistical Data Science

Received
1 November 2020

Accepted
1 January 2021

Published
10 February 2021

Abstract

We consider a continuous outcome subject to nonresponse and a fully observed covariate. We propose a spline proxy pattern-mixture model (S-PPMA), an extension of the proxy pattern-mixture model (PPMA) (Andridge and Little, 2011), to estimate the mean of the outcome under varying assumptions about nonresponse. S-PPMA improves the robustness of PPMA, which assumes bivariate normality between the outcome and the covariate, by modeling the relationship via a spline. Simulations indicate that S-PPMA outperforms PPMA when the data deviate from normality and are missing not at random, with minor losses of efficiency when the data are normal.

Supplementary material

Supplementary Material

Please refer to the Supplementary Material document for: 1. A detailed description of the Gibbs sampling algorithm for the penalized spline prediction. 2. Results from all six simulation scenarios, including estimates from n=100 and n=400 and where λA=λT and λA≠λT . 3. R code and workspace for the simulations.

References

Andridge RR, Little RJA (2010). A review of hot deck imputation for survey nonresponse. International Statistical Review, 78(1): 40–64.

Andridge RR, Little RJA (2011). Proxy pattern-mixture analysis for survey nonresponse. Journal of Official Statistics, 27: 153–180.

Little RJA (1993). Pattern-mixture models for multivariate incomplete data. Journal of the American Statistical Association, 88: 125–134.

Little RJA (1994). A class of pattern-mixture models for normal incomplete data. Biometrika, 81: 471–483.

Little RJA, An H (2004). Robust likelihood-based analysis of multivariate data with missing values. Statistica Sinica, 14: 949–968.

Little RJA, Rubin DB (2020). Statistical Analysis with Missing Data. Wiley, Third Edition.

Pfeffermann D, Sikov A (2011). Imputation and estimation under nonignorable nonresponse in household surveys with missing covariate information. Journal of Official Statistics, 27: 181–209.

Rubin DB (1976). Inference and missing data. Biometrika, 63: 581–592.

Schouten B (2007). A selection strategy for weighting variables under a not-missing-at-random assumption. Journal of Official Statistics, 23: 51–68.

Yang Y, Little RJA (2015). A comparison of doubly robust estimators of the mean with missing data. Journal of Statistical Computation and Simulation, 85: 3383–3403.

Zhang G, Little RJA (2009). Extensions of the penalized spline of propensity prediction method of imputation. Biometrics, 65(3): 911–918.

This is a free to read article.

Keywords

missing data missing not at random nonignorable nonresponse nonresponse bias

Metrics

since February 2021

1067

Article info
views

613

PDF
downloads

RSS

Authors

Abstract

Supplementary material

References

Export citation

Copy and paste formatted citation

Download citation in file