Multiple Imputation for Multivariate Missing-Data Problems: A Data Analyst's Perspective.

By Joseph L. Schafer, M. Olsen. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

BoostClean: Automated Error Detection and Repair for Machine Learning (2017, background): “While more sophisticated statistical imputation techniques exist [45], they all have the same fundamental problem that the value imputation is divorced from the down-stream classiﬁer’s predictive accuracy.”
Online impulse buying of tourism products: The role of web site personality, utilitarian and hedonic web browsing (2016, unknown): “Of several missing value remedies, multiple imputation considered as effective and reliable (Schafer and Olsen, 1998; Rezaei and Ghodsi, 2014).”
Solutions for Missing Data in Structural Equation Modeling (2006, unknown): “Schafer and Olsen (1998) note that multiple imputation methods resemble other methods of ad hoc case deletion because it addresses the missing-data issue at the beginning, before substantive analyses are run.”
Assessing Covariates of Adolescent Delinquency Trajectories: A Latent Growth Mixture Modeling Approach (2004, methodology): “Following the recommendations made by Graham and Hofer (2000) and Schafer and Olsen (1998) , the chi-square tests reported in Table II were performed on the basis of the EM covariance matrix (EM converged after 31 iterations), whereas the multinomial logistic”
Multiple Imputation of Multilevel Missing Data—Rigor Versus Simplicity (2015, methodology): “A similar approach has been suggested by Schafer and Olsen (1998), but see Yucel, He, and Zaslavsky (2008) for potential problems with this imputation strategy.”
The persistence of distance? The impact of technology on MNE motivations for foreign investment (2005, methodology): “This approach is based on the assumption that the missing values have a similar distribution to the non-missing values, an assumption commonly made when estimating missing values (e.g., Schafer and Olsen, 1998).”
Principled missing data methods for researchers (2013, methodology): “As the name suggests, FIML obtains parameter estimates by maximizing the likelihood function of the incomplete data.”
Treatment Of Missing Data At The Second Level Of Hierarchical Linear Models (2003, methodology): “Schafer and Olsen (1998) cautioned that the model used to impute data sets should contain all of the variables to be used in analyses.”