The multiple imputation was developed as a general method for inference with missing data. Instead replacing the missing observation with a single value, multiple imputation method replaces each missing value with multiple plausible values. PROC MI in SAS creates multiply imputed data sets for incomplete multivariate data. This study reviews multiple imputation as an analytic strategy for missi...

THE ANALYSIS OF MISSING DATA IN PUBLIC USE SURVEY DATABASES: A SURVEY OF STATISTICAL METHODS Ping Xu November 20, 2004 Missing data is very common in survey research. However, currently few guidelines exist with regard to the diagnosis and remedy to missing data in survey research. The goal of the thesis was to investigate properties and effects of three selected missing data handling technique...

Multiple imputation is increasingly recommended in epidemiology to adjust for the bias and loss of information that may occur in analyses restricted to study participants with complete data (‘‘complete-case analyses’’). However, little guidance is available on applying the method, including which variables to include in the imputation model and the number of imputations needed. Here, the author...

BACKGROUND Multiple imputation is a popular approach to handling missing data in medical research, yet little is known about its applicability for estimating the relative risk. Standard methods for imputing incomplete binary outcomes involve logistic regression or an assumption of multivariate normality, whereas relative risks are typically estimated using log binomial models. It is unclear whe...

There are many competing computational algorithms in multiple imputation. To this date, however, it is unknown which of these algorithms outperforms the others under what circumstances. In this paper, we describe the mechanisms of various multiple imputation algorithms and compare their performance in a variety of situations to determine which algorithm is best suited to the imputation of missi...

UNLABELLED BACKGROUND Multiple imputation is becoming increasingly popular for handling missing data. However, it is often implemented without adequate consideration of whether it offers any advantage over complete case analysis for the research question of interest, or whether potential gains may be offset by bias from a poorly fitting imputation model, particularly as the amount of missing...

Missing covariate data commonly occur in epidemiological and clinical research, and are often dealt with using multiple imputation. Imputation of partially observed covariates is complicated if the substantive model is non-linear (e.g. Cox proportional hazards model), or contains non-linear (e.g. squared) or interaction terms, and standard software implementations of multiple imputation may imp...

Multiple imputation (MI) inference handles missing data by imputing the values $m$ times, and then combining results from complete-data analyses. However, existing method for likelihood ratio tests (LRTs) has multiple defects: (i) combined test statistic can be negative, but its null distribution is approximated an $F$-distribution; (ii) it not invariant to re-parametrization; (iii) fails ensur...

Bayesian multiple imputation and maximum likelihood provide useful strategy for dealing with dataset including missing values. Imputation methods affect the significance of test results and the quality of estimates. In this paper, the general procedures of multiple imputation and maximum likelihood described which include the normal-based analysis of a multiple imputed dataset. A Monte Carlo si...