Inference using conditional logistic regression with missing covariates. (English) Zbl 1058.62542

Lipsitz, Stuart R.; Parzen, Michael; Ewell, Marian

Biometrics 54, No. 1, 295-303 (1998).

Summary: When there are many nuisance parameters in a logistic regression model, a popular method for eliminating these nuisance parameters is conditional logistic regression. Unfortunately, another common problem in a logistic regression analysis is missing covariate data. With many nuisance parameters to eliminate and missing covariates, many investigators exclude any subject with miss covariates and then use conditional logistic regression, often called a complete-case analysis.
We derive a modified conditional logistic regression that is appropriate with covariates that are missing at random. Performing a conditional logistic regression with only the complete cases is convenient with existing statistical packages, but it may give bias if missingness is not completely at random.

MSC:

62J12	Generalized linear models (logistic models)
62P10	Applications of statistics to biology and medical sciences; meta analysis

Keywords:

complete-case analysis; missing at random; missing completely at random; missing covariate data; nuisance parameters