臺灣博碩士論文加值系統

本篇論文主要在研究對一具有不相等平均數或變異數的獨立常態隨機
變數序列上之改變點個數的偵測. 在本篇論文中一種簡單的方法將被提出
來偵測改變點的個數, 這個方法稱之為 SC(k).在某些假設的情況下, 我
們將會發現根據我們方法所得到的估計量將會有 consistency a.s.
(almost surely) 的特性對這真正的改變點的個數, 而且這距離在被估計
的改變點位置和真正的改變點位置將會是 Op(d0*(n)).

A simple method is proposed to detect the number of
change points in a sequence of independent normal random
variables. An estimator to maximize some criterion, saySC(k),
which is to maximize the log likelihood function withsome
penalty term is used in detection. Under some mild
assumptions, the consistency of the estimator for the true
number of change points and the boundedness between the
estimated change locations and the true change locations are
obtained.