两值响应数据的转变点模型-概率论与数理统计专业论文.docxVIP

摘要出，我们用于估计转变点的方法是非常有效的。 摘要 出，我们用于估计转变点的方法是非常有效的。 关键词：两值响应模型，极大计分估计，转变点，强相合，收敛速度，指数收 敛速度 英文摘要Abstract 英文摘要 Abstract The binary-response model is an important among limited dependent variable (LDV)models，for which the range of the dependent variable is restricted to subset of the real line The binary response model has been widely studied[15]and[18]present empirical models of travel choices in which the coefficients random【22]gave an efficient semiparametric estimator for binary response models 130】and【32】analyze the problem of semiparametric estimation of a linear quantile regression model from random sample of binary response observations．In the assumption of linear a-quantile regression， z7口is the unique o—quantile of Y conditional z．Manski introduced the，maximum score’estimate of the normalized parameter vector卢+=卢川p||there．In this paper，we consider the change-point problem with the linear quantile regression assumption in binary response model，where observation sequence has quantile shift at an unknown time point A，which is called the change-point．Change-point problems frequently occur not only in industry automation，one of the fields in which there is appear early statistical research of change-point problems，but also in many natural and social fields such as economic， finance，seismology,epidemiology and computer science．Under some mild conditions we used the empirical distribution theory{which is created by Vapnik and Chervonenkis to establish a strong consistency estimation of change—point and give its convergence rate． In chapter 1，the development and research status of the change—point and binary 一“i一 英文摘要response 英文摘要 response model is briefly introduced．In Chapter 2，with the assumption of linear median regression，we established strong consistency estimation of change—point and proved the estimation tends to the real change-point A exponentially in the that for any given ￡0，there exists a constant C0 such that P(IA—Al≥￡)=O(exp{-Cn})．In Chapter 3，we extent i