Speakers
Description
Nonignorable nonresponse (NMAR) presents a persistent challenge in official survey statistics, where missingness mechanisms depending on unobserved variables can significantly bias key national indicators. The NMAR package addresses this issue by providing a comprehensive suite of modern estimation methods within a unified API, specifically designed for the complex reality of NMAR scenarios in official statistics.
The package implements following techniques: Generalized calibration with more variables in calibration than response model [1]; Empirical likelihood-based approaches [2]; Fractional imputation and nonparametric methods (Minsun Kim Riddles, Jae Kwang Kim, Jongho Im 2016)[3]; Induced logistic regression approaches (Li, P., Qin, J., and Liu, Y.)[4].
The package is under active development and is available on CRAN and Github. It enables researchers to produce defensible estimates from survey data affected by nonignorable missingness, while fostering methodological transparency in official statistics.
Acknowledgements
The NMAR package is developed under the project Towards census-like statistics for foreign-born populations funded by the National Science Centre, Poland (OPUS 20 grant no. 2020/39/B/HS4/00941).
Package website: https://ncn-foreigners.ue.poznan.pl/NMAR/
References
[1] Kott, P. S., Liao, D. (2017). Calibration Weighting for Nonresponse that is Not Missing at Random: Allowing More Calibration than Response-Model Variables. Journal of Survey Statistics and Methodology, 5(2), 159–174.
[2] Qin, J., et al. (2002). Estimation with Survey Data under Nonignorable Nonresponse or Informative Sampling. Journal of the American Statistical Association, 97(457), 193–200.
[3] Riddles, M. K., Kim, J. K., Im, J. (2016). A Propensity-score-adjustment Method for Nonignorable Nonresponse. Journal of Survey Statistics and Methodology, 4(2), 215–245.
[4] Li, P., Qin, J., and Liu, Y. (2023). Instability of Inverse Probability Weighting Methods and a Remedy for Nonignorable Missing Data. Biometrics, 79(4), 3215-3226.
If you used AI tools or services to support the preparation of this submission, please state the name and reason for using each of them.
No AI tools/services were used.
Additional Material or Paper
Package website: https://ncn-foreigners.ue.poznan.pl/NMAR/
| Keywords: Please list up to 5 keywords to help us find the right session for your contribution. | r package, nonignorable nonresponse, official statistics, computational statistics, estimation |
|---|---|
| Virtual Option | This submission is for onsite presentation only |
| Video Recording | Video sharing is fine |
| The author(s) agree(s) to take responsibility and be accountable for the contents of the submission and is/are authorized to present it. | Confirm |