BROWSE

ITEM VIEW & DOWNLOAD

Determining Changes in the Covariance Structure of Gaussian Processes

Cited 0 times inthomson ciCited 0 times inthomson ci
Title
Determining Changes in the Covariance Structure of Gaussian Processes
Other Titles
가우시안 과정의 공분산에서의 변화점 정의 및 검출
Author
Han, Jiyeon
Advisor
Kim, Kwang In
Issue Date
2020-02
Publisher
Graduate School of UNIST
Abstract
Time series data is everywhere, such as stock data in finance market, the sensor data in factories, or the temperature data in everyday life. Time series data have been studied for a long time to analyze and predict the future behavior. While it is tractable when the sequential data behave as stationary, it becomes difficult to model and predict non-stationary time series. Change point detection is a problem that identify non-stationarity which has been investigated for decades in many different names. Change point detection is a challenging problem because defining a change point decisively and objectively is difficult in nature. In this thesis we are trying to define and find a change point using hypothesis tests based on statistics. Specifically we focus on structural breaks in the covariance structure of Gaussian Processes. Further we propose an online change point detection algorithm, called Confirmatory Bayesian Online Change Point Detection, by leveraging the devised hypothesis tests into the conventional Bayesian online change point detection algorithm.
Description
Department of Computer Science and Engineering
URI
Go to Link
Appears in Collections:
EE_Theses_Master
Files in This Item:
Determining Changes in the Covariance Structure of Gaussian Processes.pdf Download

find_unist can give you direct access to the published full text of this article. (UNISTARs only)

Show full item record

qrcode

  • mendeley

    citeulike

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

MENU