We derive explicit formulae for estimation in logistic regression models where some of the covariates are missing. Our approach allows for modelling the distribution of the missing covariates either as a multivariate normal or as a multivariate t-distribution. A main advantage of this method is that it is fast and does not require the use of iterative procedures. A model selection method is derived which allows to choose among these distributions. In addition, we consider versions of Akaike’s information criterion that are based on the expectation–maximization algorithm and multiple imputation methods that have a wide applicability to model selection in likelihood models in general.
Akaike H (1973) Information theory and an extension of the maximum likelihood principle. In Petrov B and Csáki F eds. Second international symposium on information theory. Budapest: Akadémiai Kiadó, 267–81.
2.
Blackhurst DW and Schluchter MD (1989) Logistic regression with a partially observed covariate. Communication in Statistics–-Simulation, 18, 163–77.
3.
Cavanaugh JE and Shumway RH (1998) An Akaike information criterion for model selection in the presence of incomplete data. Journal of Statistical Planning and Inference, 67, 45–65.
4.
Claeskens G and Consentino F (2008) Variable selection with incomplete covariate data. Biometrics, 64, 1062–69.
5.
Consentino F and Claeskens G (2010) Order selection tests with multiply-imputed data. Computational Statistics and Data Analysis, 54, 2284–95.
6.
Dempster AP, Laird NM and Rubin DB (1977) Maximum likelihood from incomplete data via EM algorithm. Journal of the Royal Society, Series B, 39, 1–22.
7.
Gao S and Hui SL (1997) Logistic regression models with missing covariate values for complex survey data. Statistics in Medicine, 16, 2419–28.
8.
Gilks WR and Wild P (1992) Adaptive rejection sampling for Gibbs sampling. Applied Statistics, 41, 337–48.
9.
Hens N, Aerts M and Molenberghs G (2006) Model selection for incomplete and design-based samples. Statistics in Medicine, 25, 2502–20.
10.
Ibrahim JG (1990) Incomplete data in generalized linear models. Journal of the American Statistical Association, 85, 765–69.
11.
Ibrahim JG, Chen M-H and Lipsitz SR (1999a) Monte Carlo EM for missing covariates in parametric regression models. Biometrics, 55, 591–96.
12.
Ibrahim JG, Lipsitz SR and Chen M-H (1999b) Missing covariates in generalized linear models when the missing data mechanism is non-ignorable. Journal of the Royal Statistical Society, Series B, 61, 173–190.
13.
Kotz S and Nadarajah S (2004) Multivariate t distributions and their applications. Cambridge: Cambridge University Press.
14.
Li KH, Raghunathan TE and Rubin DB (1991) Large-sample significance levels from multiply imputed data using moment-based statistics and an F reference distribution. Journal of the American Statistical Association, 86, 1065–73.
15.
Lipsitz, Stuart R, Parzen M and Ewell M (1998) Inference using conditional logistic regression with missing covariates. Biometrics, 54, 295–303.
16.
Little RJA and Rubin DB (2002) Statistical analysis with missing data, 2nd edition. Wiley Series in Probability and Statistics. Hoboken, NJ: Wiley-Interscience [John Wiley & Sons].
17.
Liu C (1995) Missing data imputation using the multivariate t distribution. Journal of Multivariate Analysis, 53, 139–58.
18.
Liu C and Rubin DB (1995) ML estimation of the multivariate t distribution with unknown degrees of freedom. Statistica Sinica, 5, 19–39.
19.
Meng X-L and Rubin DB (1992) Performing likelihood ratio tests with multiply-imputed data sets. Biometrika, 79, 103–11.
20.
Raghunathan TE, Lepkowski J, Van Hoewyk J and Solenberger P (2001) A multivariate technique for multiply imputing missing values. Survey Methodology, 27, 85–95.
21.
Schomaker M, Heumann C and Toutenburg H (2007) New approaches for model selection under missing data. Technical Report, Department of Statistics, LudwigMaximilians Universitäat München.
22.
Shimodaira H (1994) A new criterion for selecting models from partially observed data. In Cheeseman P and Oldford RW (eds.). Selecting models from data: artificial intelligence and statistics IV. New York: Springer, 21–29.
23.
Yang X, Belin TR and Boscardin WJ (2005) Imputation and variable selection in linear regression models with missing covariates. Biometrics, 61, 498–506.
