Semiparametric linear transformation models for indirectly observed outcomes
- Author
- Jan De Neve (UGent) and Heidelinde Dehaene (UGent)
- Organization
- Project
- Abstract
- We propose a regression framework to analyze outcomes that are indirectly observed via one or multiple proxies. Semiparametric transformation models, including Cox proportional hazards regression, turn out to be well suited to model the association between the covariates and the unobserved outcome. By coupling this regression model to a semiparametric measurement model, we can estimate these associations without requiring calibration data and without imposing strong functional assumptions on the relationship between the unobserved outcome and its proxy. When multiple proxies are available, we propose a data-driven aggregation resulting in an improved proxy. We empirically validate the proposed methodology in a simulation study, revealing good finite sample properties, especially when multiple proxies are aggregated. The methods are demonstrated on two case studies.
- Keywords
- Statistics and Probability, Epidemiology, indicators, measurement error, probabilistic index model, proxy
Downloads
-
(...).pdf
- full text (Published version)
- |
- UGent only
- |
- |
- 1.13 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-8693148
- MLA
- De Neve, Jan, and Heidelinde Dehaene. “Semiparametric Linear Transformation Models for Indirectly Observed Outcomes.” STATISTICS IN MEDICINE, vol. 40, no. 9, 2021, pp. 2286–303, doi:10.1002/sim.8903.
- APA
- De Neve, J., & Dehaene, H. (2021). Semiparametric linear transformation models for indirectly observed outcomes. STATISTICS IN MEDICINE, 40(9), 2286–2303. https://doi.org/10.1002/sim.8903
- Chicago author-date
- De Neve, Jan, and Heidelinde Dehaene. 2021. “Semiparametric Linear Transformation Models for Indirectly Observed Outcomes.” STATISTICS IN MEDICINE 40 (9): 2286–2303. https://doi.org/10.1002/sim.8903.
- Chicago author-date (all authors)
- De Neve, Jan, and Heidelinde Dehaene. 2021. “Semiparametric Linear Transformation Models for Indirectly Observed Outcomes.” STATISTICS IN MEDICINE 40 (9): 2286–2303. doi:10.1002/sim.8903.
- Vancouver
- 1.De Neve J, Dehaene H. Semiparametric linear transformation models for indirectly observed outcomes. STATISTICS IN MEDICINE. 2021;40(9):2286–303.
- IEEE
- [1]J. De Neve and H. Dehaene, “Semiparametric linear transformation models for indirectly observed outcomes,” STATISTICS IN MEDICINE, vol. 40, no. 9, pp. 2286–2303, 2021.
@article{8693148, abstract = {{We propose a regression framework to analyze outcomes that are indirectly observed via one or multiple proxies. Semiparametric transformation models, including Cox proportional hazards regression, turn out to be well suited to model the association between the covariates and the unobserved outcome. By coupling this regression model to a semiparametric measurement model, we can estimate these associations without requiring calibration data and without imposing strong functional assumptions on the relationship between the unobserved outcome and its proxy. When multiple proxies are available, we propose a data-driven aggregation resulting in an improved proxy. We empirically validate the proposed methodology in a simulation study, revealing good finite sample properties, especially when multiple proxies are aggregated. The methods are demonstrated on two case studies.}}, author = {{De Neve, Jan and Dehaene, Heidelinde}}, issn = {{0277-6715}}, journal = {{STATISTICS IN MEDICINE}}, keywords = {{Statistics and Probability,Epidemiology,indicators,measurement error,probabilistic index model,proxy}}, language = {{eng}}, number = {{9}}, pages = {{2286--2303}}, title = {{Semiparametric linear transformation models for indirectly observed outcomes}}, url = {{http://doi.org/10.1002/sim.8903}}, volume = {{40}}, year = {{2021}}, }
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: