Spatio-temporal prediction modeling of clusters of influenza cases

• Author / Creator

  Qiu, Weiyu

• Timely, accurate predictions of potential influenza epidemics are essential for healthcare providers and policy makers as the epidemics can result in heavy demands for health services. Current statistical modeling of surveillance data has limited prediction abilities and often fails to respond effectively to the outbreaks. The first part of this thesis, a collaboration with Alberta Health Services, aims at predicting clusters of influenza cases in Edmonton weeks in advance, using real-time data collected from emergency-department visits by Alberta Real Time Syndromic Surveillance Net. The 2004-2009 data are analyzed by spatio-temporal modeling and predictions are cross-validated. In the second part of this thesis, a related theoretical work on multivariate modeling, with spatio-temporal modeling as a potential application, is presented, proving that every conditional second moment is linear in the empirical second moment of the conditioning vector if and only if the distribution belongs to the multivariate Pearson type VII family.

• Subjects / Keywords

  • Generalized linear mixed model
  • Cross-validation
  • Influenza prediction
  • Multivariate Pearson Type VII family
  • Pseudo-likelihood

• Graduation date

  Fall 2011

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R3G94T

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.