科研之友
微信
新浪微博
Facebook
分享链接
摘要
This paper studies rescaled images, under exp(mu)(-1) of the sample Frechet means of i.i.d. random variables {X-k vertical bar k >= 1} with Frechet mean mu on a Riemannian manifold. We show that, with appropriate scaling, these images converge weakly to a diffusion process. Similar to the Euclidean case, this limiting diffusion is a Brownian motion up to a linear transformation. However, in addition to the covariance structure of exp(mu)(-1) (X-1), this linear transformation also depends on the global Riemannian structure of the manifold.
- 出版日期2015-12
