Statistical Model Of Mutation Detection In The Weighted Residual Method

Change-point problems have originally arisen in the context of many natural and social fields such as quality control, economic, finance, seismology, epidemiology and computer science. The researches on the thesis are very valuable for statistical theories and applications.This thesis is divided into three chapters.In the fist chapter, we introduced the development and research status of change-point.In chapter two, the MOSUM method is proposed based on weighted residual moving sum in order to detect change-points in IV models. Under some non-orthogonality condition, the test has non-trival power whenδ= 1/2 and will be consistent when 1≤δ≤1/2.In the third chapter, the detection of change-points in IV models is proposed with the help of Cramer-Von Mises method which is based on weighted residual partial sum. Under some non-orthogonality condition, the test has non-trival power whenδ= 1/2 and will be consistent when 1≤δ≤1/2.We investigate our distribution theory with a Monte Carlo simulation which indicates the applicability of our methods.