Monitoring Parameter Change in Autocorrelated Logistic Regression

Monitoring changes in health care performances, financial markets, and industrial processes has recently gained momentum due to increased capability of computers and the availability of real-time data collection and storage software. As a consequence, there has been a growing demand in developing statistically rigorous methodologies for monitoring and change-point detection. In many practical situations, the data being monitored for the purpose of detecting changes, present serial correlations. Hussein (2011) is currently working on a new statistical procedure for monitoring changes in the coefficients of logistic regression model with AR( p)-type structure. The objective of this thesis is (a) to use Monte Carlo experiments to evaluate the average stopping times, probability of false alarm, and power of the proposed procedure; (b) to illustrate the usefulness of the method by using an IBM stock transactions data as well as data on rainfall.

Name thesis:degree_name

M.Sc.

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Mathematics and Statistics

Grantor

University of Windsor

Year dc:date.issued

2011

Author dc:creator

• Khaleghei Ghosheh Balagh, Akram

Advisor dc:contributor.advisor

• Hussein, Abdulkadir Ali

Contributors dc:contributor

• olsenc@uwindsor.ca

Licence

• http://creativecommons.org/licenses/by-nc-nd/4.0/

Language dc:language.iso

en_CA

Handle dc:identifier.uri

https://hdl.handle.net/20.500.14776/8214

OAI identifier oai:identifier

oai:uwindsor.scholaris.ca:20.500.14776/8214