106/11/20 李百靈 教授(淡江大學統計學系)
主講人: 李百靈 教授(淡江大學統計學系)
題 目：Supervised classification of functional data using subspace projection
We propose a supervised classification method for functional data, which takes the additional covariate information into account. The response functions in each class are embedded in a cluster subspace spanned by a covariate-adjusted mean function and a set of eigenfunctions of the covariance kernel based on the Karhunen-Loève expansion. Under the assumption of distinct subspaces of each class, a newly observed function is classified into the best-predicted class by minimizing the distance between the observation and its projection onto the class subspaces among all classes. The covariate adjustment is useful for functional classification, especially when the covariate effects on the mean functions are significantly different among the classes. We demonstrate the proposed classification method and its performance through simulation studies and real data application.
日 期：106年11月20日(星期一) 16:00~17:00
茶 會： 15:30~16:00數學館四樓409室舉行