Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/5005
Title: | Model selection for regression analyses with missing data | Authors: | AERTS, Marc HENS, Niel MOLENBERGHS, Geert |
Issue Date: | 2004 | Publisher: | Firenze University Press | Source: | Biggeri, Annibale; Dreassi, Emanuela; Lagazio, Corrado; Marchi, Marco (Ed.). Proceedings of the 19th International Workshop on Statistical Modelling. p. 43-47. | Abstract: | The Akaiki Information Criterion, AIC, is one of the leading selection methods for regression models. In case of partially missing covariates with missingness probability depending on the response, regression estimates based on the so-called complete cases are known to be biased. In this contribution it is shown that model selection using AIC-values based on the complete cases can lead to the choice of wrong or less optimal models. In analogy with the weighted Horvitz-Thompson estimator, we propose a weighted version of AIC. It is shown that this weighted AIC criterion improves model choices. | Keywords: | Akaiki information criterion; missing data; model selection; weighted likelihood | Document URI: | http://hdl.handle.net/1942/5005 | Link to publication/dataset: | http://www.statmod.org/files/proceedings/iwsm2004_proceedings.pdf | ISBN: | 8884531934 | Rights: | (C) 2004 Firenze University Press | Category: | C2 | Type: | Proceedings Paper |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
iwsm2004_proceedings-63-67.pdf | 67.15 kB | Adobe PDF | View/Open |
Page view(s)
18
checked on Jun 7, 2023
Download(s)
6
checked on Jun 7, 2023
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.