Visualizing and measuring agreement between two long ranked lists

3^rd International Conference & Exhibition on Biometrics & Biostatistics

October 20-21, 2014 DoubleTree by Hilton Baltimore - BWI Airport, USA

Srinath Sampath and Joseph S Verducci

Scientific Tracks Abstracts: J Biom Biostat

Abstract :

The stage-wise agreement between two long ranked lists is characterized by adapting Fligner and Verducci?s (1988) multistage ranking model into the moving average maximum likelihood estimator (MAMLE), a locally smooth estimator that is used to measure the strength of association between the lists and to provide a stopping rule to detect the stage where the signal ultimately degenerates into noise. This MAMLE stopping rule is further applied to bivariate data arranged in tau-path order to measure the length of the relationship instrongly associated subsets of gene expression and chemical potency data from the National Cancer Institute.

Biography :

Srinath Sampath has a PhD in Statistics from The Ohio State University, is an Associate of the Society of Actuaries, and a CFA Charter holder. He is a portfolio manager and trader at Hamilton Capital Management, a registered investment advisor in Columbus, Ohio, where his team manages $1.4 Billion in assets for clients all over the United States. Srinath?s current interests include model development to uncover value in various asset classes in equity and fixed income markets. His research interests include ranking algorithms, nonparametric statistics, time series modeling, and statistical learning methods in quantitative finance.