Home  >>  Archives  >>  Volume 9 Number 3  >>  st0170

The Stata Journal
Volume 9 Number 3: pp. 388-397



Subscribe to the Stata Journal
cover

Improved degrees of freedom for multivariate significance tests obtained from multiply imputed, small-sample data

Yulia V. Marchenko
StataCorp
College Station, TX
ymarchenko@stata.com
Jerome P. Reiter
Department of Statistical Science
Duke University
Durham, NC
jerry@stat.duke.edu
Abstract.  We propose improvements to existing degrees of freedom used for significance testing of multivariate hypotheses in small samples when missing data are handled using multiple imputation. The improvements are for 1) tests based on unrestricted fractions of missing information and 2) tests based on equal fractions of missing information with M(p−1) ≤ 4, where M is the number of imputations and p is the number of tested parameters. Using the mi command available as of Stata 11, we demonstrate via simulation that using these adjustments can result in a more sensible degrees of freedom (and hence closer-to-nominal rejection rates) than existing degrees of freedom.
Terms of use     View this article (PDF)

View all articles by these authors: Yulia V. Marchenko, Jerome P. Reiter

View all articles with these keywords: multiple imputation, degrees of freedom, sample, missing, testing, multivariate

Download citation: BibTeX  RIS

Download citation and abstract: BibTeX  RIS