Fisher's Information Matrix: A new way to do dimension reduction in semiparametric multivariate analysis
Bruce Lindsay (Penn State University)
Friday 22nd March, 2013 15:00-16:00 Maths 204
We will show that the Fisher's information matrix has great potential for the analysis of multivariate data. To date we have shown that the eigenanalysis of this matrix, and other matrices derived from it, can be employed for dimension reduction in many standard multivariate analyses. include projection pursuit, independent components analysis, graphical model fitting, covariate dimension reduction, and population discrimination. We will give some theory and some examples.