 Analysis of high-dimensional repeated measures designs: The one sample caseM. Rauf Ahmad, C. Werner and E. Brunner Computational Statistics & Data Analysis, 2008, vol. 53, issue 2, 416-427 Abstract: A one sample statistic is derived for the analysis of repeated measures design when the data are multivariate normal and the dimension, d, can be large compared to the sample size, n, i.e. d>n. Quadratic and bilinear forms are used to define the statistic based on Box's approximation [Box, G.E.P., 1954. Some theorems on quadratic forms applied in the study of analysis of variance problems I: Effect of inequality of variance in the one-way classification. Annals of Mathematical Statistics 25 (2), 290-302]. The statistic has an approximate distribution, even for moderately large n. One of the main advantages of the statistic is that it can be used both for unstructured and factorially structured repeated measures designs. In the asymptotic derivations, it is assumed that n-->[infinity] while d remains finite and fixed. However, it is demonstrated through simulations that for n as small as 10, the new statistic very closely approximates the target distribution, unaffected by even large values of . The application is illustrated using a sleep lab example with . Date: 2008 Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:eee:csdana:v:53:y:2008:i:2:p:416-427 Access Statistics for this article Computational Statistics & Data Analysis is currently edited by S.P. Azen More articles in Computational Statistics & Data Analysis from Elsevier |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.