MODELING THE LABELING INDEX DISTRIBUTION - AN APPLICATION OF FUNCTIONAL DATA-ANALYSIS

This article presents exploratory data analytic methodology for visual izing and summarizing data that can be represented as individual-speci fic curves. We propose a simplified form of functional data analysis. A nonparametric scatterplot smooth is applied to each individual's dat a, followed by a principal components analysis of the smoothed data. W e then display the individual smooth curves in an array organized by p rincipal component scores. The display suggests interpretable summary measures. The methodology is applied to the measurement of proliferati ve activity, a biomarker for colon cancer risk. We use the summary mea sures in the analysis of a pilot study clinical trial.