Hippocampus Shape Analysis of Clinically Depressed Twin Populations Using a Modified Mann-Whitney-Wilcoxon Statistic

• Activity Number: 275
• Type: Contributed
• Date/Time: Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Nikhil Ram Mohan*+ and Carey E. Priebe and Youngser Park and Majnu John
• Companies: Johns Hopkins University and Johns Hopkins University and Johns Hopkins University and Weill Medical College of Cornell University
• Keywords: Mann-Whitney-Wilcoxon ; Pitman Asymptotic Efficacy ; Major Depressive Disorder ; Hippocampus

Abstract

The Mann-Whitney-Wilcoxon (MWW) test statistic, while distribution-free, suffers from a loss of efficacy for certain underlying distributions. In this manuscript, we instead use a data-adaptive weighted generalized Mann-Whitney-Wilcoxon (AWGMWW) test statistic, one that is optimal in the Pitman Asymptotic Efficacy (PAE) sense, to discern differences in hippocampus shape among twin populations with or without Major Depressive Disorder (MDD). We verify, based on a previous study using the MWW statistic, that a high-risk group is more similar to the control group than the depressed group in terms of hippocampus shape. In addition, we show that the control group is more similar to the high-risk group than the depressed group---a distinction that could not be made in the preceding study. This suggests that the AWGMWW statistic is more powerful in this application.