Testing of coarsening mechanisms: Coarsening at random versus subgroup independence

Cattaneo, Marco E. G. V.; Augustin, Thomas; Plass, Julia; Schollmeyer, Georg

Abstract

Abstract Since coarse(ned) data naturally induce set-valued estimators, analysts often assume coarsening at random (CAR) to force them to be single-valued. Using the PASS data as an example, we re-illustrate the impossibility to test CAR and contrast it to another type of uninformative coarsening called subgroup independence (SI). It turns out that SI is testable.

Citation

Cattaneo, M. E. G. V., Augustin, T., Plass, J., & Schollmeyer, G. (2017). Testing of coarsening mechanisms: Coarsening at random versus subgroup independence. Advances in Intelligent Systems and Computing, 456, 415-422. https://doi.org/10.1007/978-3-319-42972-4_51

• Journal Article Type: Article
• Acceptance Date: Jun 1, 2016
• Online Publication Date: Jul 30, 2016
• Publication Date: Jan 1, 2017
• Deposit Date: Jun 17, 2016
• Journal: Soft methods for data science
• Print ISSN: 2194-5357
• Electronic ISSN: 2194-5365
• Publisher: Springer Verlag
• Peer Reviewed: Not Peer Reviewed
• Volume: 456
• Pages: 415-422
• Book Title: Advances in Intelligent Systems and Computing; Soft Methods for Data Science
• ISBN: 978-3-319-42971-7
• DOI: https://doi.org/10.1007/978-3-319-42972-4_51
• Keywords: Coarse data, Missing data, Coarsening at random (CAR), Hypothesis testing, Likelihood-ratio test
• Public URL: https://hull-repository.worktribe.com/output/439796