Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/oxford-university-press/a-conditional-model-for-incomplete-covariates-in-parametric-regression-5AA6f060I5
References (9)
-
G. Falkson, S. Lipsitz, E. Borden, I. Simson, D. Haller (1995)
Hepatocellular Carcinoma: An ECOG Randomized Phase II Study of Beta‐Interferon and MenagorilAmerican Journal of Clinical Oncology, 18
-
Martin Gibson, R. Little, D. Rubin (1989)
Statistical Analysis with Missing Data.The Statistician, 38
-
S. Fienberg, W. Mason (1979)
Identification and estimation of age-period-cohort models in the analysis of discrete archival dataSociological Methodology, 10
-
D. Rubin (1975)
INFERENCE AND MISSING DATAPsychometrika, 1975
-
T. Louis (1982)
Finding the Observed Information Matrix When Using the EM AlgorithmJournal of the royal statistical society series b-methodological, 44
-
G. Falkson, A. Cnaan, I. Simson, Y. Dayal, H. Falkson, T. Smith, D. Haller (1990)
A Randomized Phase II Study of Acivicin and 4'Deoxydoxorubicin in Patients with Hepatocellular Carcinoma in an Eastern Cooperative Oncology Group StudyAmerican Journal of Clinical Oncology, 13
-
J. Ibrahim (1990)
Incomplete Data in Generalized Linear ModelsJournal of the American Statistical Association, 85
-
R. Little, M. Schluchter (1985)
Maximum likelihood estimation for mixed continuous and categorical data with missing valuesBiometrika, 72
-
P. McCullagh (1980)
Regression Models for Ordinal DataJournal of the royal statistical society series b-methodological, 42
- Publisher
- Oxford University Press
- Copyright
- © 1996 Biometrika Trust
- ISSN
- 0006-3444
- eISSN
- 1464-3510
- DOI
- 10.1093/biomet/83.4.916
- Publisher site
- See Article on Publisher Site
Abstract
Abstract Incomplete covariate data arise in many data sets. When the missing covariates are categorical, a useful technique for obtaining parameter estimates is the EM algorithm by the method of weights proposed in Ibrahim (1990). This method requires the estimation of many nuisance parameters for the distribution of the covariates. Unfortunately, in data sets when the percentage of missing data is high, and the missing covariate patterns are highly non-monotone, the estimates of the nuisance parameters can lead to highly unstable estimates of the parameters of interest. We propose a conditional model for the covariate distribution that has several modelling advantages for the E-step and provides a reduction in the number of nuisance parameters, thus providing more stable estimates in finite samples. We present a clinical trials example with six covariates, five of which have some missing values. © 1996 Biometrika Trust
Journal
Biometrika – Oxford University Press
Published: Dec 1, 1996