Home > Publications . Search All . Browse All . Country . Browse PSC Pubs . PSC Report Series

PSC In The News

RSS Feed icon

Shaefer says the details matter in child tax reform

Prescott says Michigan's restrictive sex offender law hurts social reentry

Yang's work suggests Hurricane Maria will prompt increased migration from Puerto Rico to US

More News


U-M is the only public and non-coastal university on Forbes' top-10 list for billionaire production

ASA President Bonilla-Silva takes exception with Chief Justice Roberts' 'gobbledygook' jab

Nobel laureate Angus Deaton, David Lam, and colleagues discuss global poverty, 10/5, 4pm

James Jackson named inaugural recipient of U-M Diversity Scholar Career Award

More Highlights

Next Brown Bag

Mon, Oct 23, 2017, noon: Carol Shiue, "Social Mobility in China, 1300-1800"

On the Performance of Sequential Regression Multiple Imputation Methods with Non Normal Error Distributions

Publication Abstract

He, Y.L., and Trivellore Raghunathan. 2009. "On the Performance of Sequential Regression Multiple Imputation Methods with Non Normal Error Distributions." Communications in Statistics: Simulation and Computation, 38(4): 856-883.

Sequential regression multiple imputation has emerged as a popular approach for handling incomplete data with complex features. In this approach, imputations for each missing variable are produced based on a regression model using other variables as predictors in a cyclic manner. Normality assumption is frequently imposed for the error distributions in the conditional regression models for continuous variables, despite that it rarely holds in real scenarios. We use a simulation study to investigate the performance of several sequential regression imputation methods when the error distribution is flat or heavy tailed. The methods evaluated include the sequential normal imputation and its several extensions which adjust for non normal error terms. The results show that all methods perform well for estimating the marginal mean and proportion, as well as the regression coefficient when the error distribution is flat or moderately heavy tailed. When the error distribution is strongly heavy tailed, all methods retain their good performances for the mean and the adjusted methods have robust performances for the proportion; but all methods can have poor performances for the regression coefficient because they cannot accommodate the extreme values well. We caution against the mechanical use of sequential regression imputation without model checking and diagnostics.

DOI:10.1080/03610910802677191 (Full Text)

Browse | Search : All Pubs | Next