电竞投注

09月18日 南开大学王光辉博士学术报告

发布时间: 2020-09-15   浏览次数: 15

电竞投注bao gao ren: wangguanghui boshi(nankaidaxue)

电竞投注baogaotimu: change-point detection by order-preserving-splitting methods  

电竞投注baogaoshijian: 2020nian9yue18ri(zhouwuxiawu6: 00 )

baogaodidian: jiangsushifandaxueshuxueyutongjixueyuanxueshubaogaoting(jingyuanlou1506shi)

wangguanghuiboshijianjie:

2018nianboshibiyeyunankaidaxuehourenzhiyunankaidaxuetongjiyushujukexuexueyuan。yanjiufangxiangweibiandianfenxihegaoweishujutongjituiduan。zaiann. statist.hej. multivariate anal.dengtongjixueqikanfabiaoduopianlunwen。

baogaozhaiyao: in multiple change-point analysis, one of the major challenges is to estimate the number of change-points. most existing approaches attempt to minimize a schwarz information criterion which balances a term quantifying model fit with a penalization term accounting for model complexity that increases with the number of change-points and limits overfitting. however, different penalization terms are required to adapt to different contexts of multiple change-point problems and the optimal penalization magnitude usually varies from the model and error distribution. we propose a data-driven selection criterion that is applicable to most kinds of popular change-point detection methods, including binary segmentation and optimal partitioning algorithms. the key idea is to select the number of change-points that minimizes the squared prediction error, which measures the fit of a specified model for a new sample. we develop a cross-validation estimation scheme based on an order-preserved sample-splitting strategy, and establish its asymptotic selection consistency under some mild conditions. effectiveness of the proposed selection criterion is demonstrated on a variety of numerical experiments and real-data examples.


电竞投注-电竞赛事投注平台 电竞投注-电竞赛事投注平台 电竞投注-电竞赛事投注平台 电竞投注-电竞赛事投注平台 电竞投注-电竞赛事投注平台 电竞投注-电竞赛事投注平台 电竞投注-电竞赛事投注平台