PhD Biostatistics

Comprehensive Examination

Students generally write the comprehensive examination after completion of the first year of study, and no later then the end of the second year. The examination, which is offered in the late summer, consists of a theoretical as well as a practical component. The theoretical component, involves two in-class exams of 5-hours each, covers the areas of probability, mathematical statistics, survival analysis, categorical data analysis, generalized liner models, semi- and non-parametric methods, and statistical epidemiology. For the practical component the student is given one week to submit a report which summarizes the statistical analysis of at least one dataset.

Supervision

PhD students must be supervised by a faculty member who has a full appointment with the School of Graduate Studies. Faculty outside of the core Biostatistics Division, such as faculty with cross-appointments, are eligible to supervise, but only if they are actively involved in and very familiar with the graduate program. In addition to a primary supervisor, a PhD Thesis Committee must have at least two other faculty members, at least one of whom holds Full SGS appointment. The other member may have Associate SGS status. These additional members are usually added as the thesis protocol is developed.

Regular contact between the student and the Thesis Committee is essential, and a formal annual review of progress is mandatory. Students and supervisors should refer to the School of Graduate Studies handbook Graduate Supervision: Guidelines for Students, Faculty and Administrators as well as to the policies and procedures of the Department of Public Health Sciences.

PhD Thesis

The thesis typically comprises four components:

• The reading and deep understanding of a set of journal papers on a particular subject in applied statistics
• The application of algebra and other mathematical techniques to advance an idea
• The use and understanding of simulation methods as an example of a well designed experiment to characterize the small sample properties of methods for which asymptotic or large sample theory has been developed
• The application of the method to a real dataset

The material in the thesis shall be original, or a non-trivial and in-depth application of advanced methods. The results of a PhD thesis must be of the calibre suitable for the publication of at least one article in an academic journal and for presentation at national and international meetings of statistics. The actual publication of the results is not a requirement prior to any defense of the thesis. The student must successfully defend the thesis in a departmental oral defense before being permitted to the final senate oral.

Other Information

Coming soon!

Links

Department of Computer Science http://www.cs.toronto.edu/DCS/index.html

Department of Statistics http://www.utstat.utoronto.ca

Biostatistics Main Page

PhD

• Description & Objectives
• Admission
• Course Requirements
• Comprehensive
• Supervision
• Thesis
• Other Information