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

PSC In The News

RSS Feed icon

Surprising findings on what influences unintended pregnancy from Wise, Geronimus and Smock

Recommendations on how to reduce discrimination resulting from ban-the-box policies cite Starr's work

Brian Jacob on NAEP scores: "Michigan is the only state in the country where proficiency rates have actually declined over time."

More News

Highlights

Call for papers: Conference on computational social science, April 2017, U-M

Sioban Harlow honored with 2017 Sarah Goddard Power Award for commitment to women's health

Post-doc fellowship in computational social science for summer or fall 2017, U-Penn

ICPSR Summer Program scholarships to support training in statistics, quantitative methods, research design, and data analysis

More Highlights

Next Brown Bag

Mon, March 13, 2017, noon:
Rachel Best

Multiple imputation using multivariate gh transformations

Publication Abstract

He, Y., and Trivellore Raghunathan. 2012. "Multiple imputation using multivariate gh transformations." Journal of Applied Statistics, 39(10): 2177-2198.

Multiple imputation has emerged as a popular approach to handling data sets with missing values. For incomplete continuous variables, imputations are usually produced using multivariate normal models. However, this approach might be problematic for variables with a strong non-normal shape, as it would generate imputations incoherent with actual distributions and thus lead to incorrect inferences. For non-normal data, we consider a multivariate extension of Tukey's gh distribution/transformation [38] to accommodate skewness and/or kurtosis and capture the correlation among the variables. We propose an algorithm to fit the incomplete data with the model and generate imputations. We apply the method to a national data set for hospital performance on several standard quality measures, which are highly skewed to the left and substantially correlated with each other. We use Monte Carlo studies to assess the performance of the proposed approach. We discuss possible generalizations and give some advices to practitioners on how to handle non-normal incomplete data.

DOI:10.1080/02664763.2012.702268 (Full Text)

Browse | Search : All Pubs | Next