STAT 432/532: Survival Analysis

Ivan Mizera (contact). Office hours: Tuesday and Thursday 2:30-4:00(pm)
TR 12:30-1:50(pm) T BW 2
  • STAT 432: Survival models, model estimation from complete and incomplete data samples, parametric survival models with concomitant variables, estimation of life tables from general population data. Prerequisite: STAT 372 and 378.
  • STAT 532: Survival and hazard functions, censoring, truncation. Non-parametric, parametric and semi-parametric approaches to survival analysis including Kaplan-Meier estimation and Cox's proportional hazards model. Prerequisite: STAT 372 or consent of Department.
    To acquire proper understanding and working knowledge of computable methods used for the analysis of time-to-event (survival, failure) data - and for STAT 532 graduates, also some knowledge about their deeper mathematical underpinning.
  • Survival Analysis Using S: Analysis of Time-to-Event Data Mara Tableman, Jong Sung Kim, Chapman & Hall/CRC, 2003.
  • Applied Survival Analysis: Regression Modeling of Time to Event Data, 2nd Edition, by David W. Hosmer Jr., Stanley Lemeshow, Susanne May, Wiley, 2008.
  • Applied Survival Analysis Using R, by Dirk F. Moore, Springer, 2016.
  • Survival and event history analysis: a process point of view, by Odd O. Aalen, Ornulf Borgan, Hakon K. Gjessing, Springer, 2008.
  • Modeling Survival Data: Extending the Cox Model, by Terry N. Therneau and Patricia M. Grambsch, Springer, 2000.
  • Specifics of survival data and objectives of their analyses. Censoring, truncation, their types and aspects. Extensions.
  • Basic quantities for discrete (mortality tables) and general data (survival function, hazard functions).
  • Likelihood methods. Kaplan-Meier, Nelson(-Aalen), Fleming-Harrington estimators. Greenwood formula.
  • Parametric modeling: relevant distributions and their properties, estimatiom, testing.
  • Inclusion of covariates, generalized linear models.
  • Accelerated time model.
  • Cox model of proportional hazards.
  • Model checking and diagnostics, residuals.
  • Frailty and competing risks models.
    The raw cumulative score will be computed from the performance in
  • STAT 432: homework assignments (40%), midterm examination (22%), and the final examination (38%).
  • STAT 532: homework assignments (30%), course project and its presentation (10%) midterm examination (22%) , and the final examination (38%).
  • The scores thus obtained will determine the final grade in the following manner: those who achieve 91% or more will get A; 76% or more qualifies for B; 61% or more guarantees C, and 51% or more the passing grade D. Fine distinctions (+ and -) will be determined from the relative position of the student's score; also, one or several of the above cutpoints may be lowered if circumstances render it necessary.
    Grades are unofficial until approved by the Department and/or Faculty offering the course. Samples of past or representative evaluative course material will be made available through the course web page.
    Students are supposed to attend all classes and actively participate in discussions and other class activities.
    About 4-6 assignments are planned: they will contain both theoretical and computational problems. The latter (not necessarily the former, however) should be printed, all necessary pictures embedded in the text. Code should be included as an appendix.
    Examination are closed-book, closed-notes, without any access to the internet; to prevent misunderstandings, cellphones, laptops and similar electronic devices are not to be brought to the exams. The questions will concentrate on the correct understanding of the material, testing it on suitable problems; therefore it is highly recommended that students work independently on the homework assignments. The examinations will not involve any problems requiring computation using software or computer; however, computational assignments are still important for gaining the proper understanding of the course material, and it is highly recommended that students also work independently on those. Examination questions may (and for the final they certainly will) be different for STAT 432 and STAT 532.
    The software used in the course is R, a freely available language and environment for statistical computing and graphics. It is available for pretty much every computational platform (Windows, OS X, Linux) in general use, there is a host of supplemental documentation and tutorials, most of them freely available on the internet, and its extensions, in the form of contributed packages, cover most of the existing statistical methodology.
  • Midterm exam: Tuesday, March 6, 2018, class time (12:30-1:50) and venue (T BW 2).
  • Transparencies: 1 | 2 | 3 | 4 | 5 | 6 | 7 |
  • Assignments: 1 | 2 | 3 |
  • Data: wm2000.d | life2014.d | bonemarr.d |
  • Sample problems:
    There are neither deferred term exams nor extended deadlines for the homework. A student who cannot complete the assignment or misses a term examination because of an incapacitating illness, severe domestic affliction or other compelling reasons can apply for an excused absence. To apply for an excused absence, a student must inform the instructor within two working days following the scheduled date of the term work or term exam missed, or as soon as the student is able, having regard to the circumstances underlying the absence. In all cases, instructors may request adequate documentation to substantiate the reason for the absence at their discretion. If the absence is excused, the missed component weight will be deferred to the final exam.
    For an excused absence where the cause is religious belief, a student must contact the instructor within two weeks of the start of Fall classes to request accommodation for the term (including the final exam, where relevant). Instructors may request adequate documentation to substantiate the student request.
    A student who cannot write the final examination due to incapacitating illness, severe domestic affliction or other compelling reasons can apply for a deferred final examination. Students who failed at the start of term to request exam accommodations for religious beliefs are expected to follow the normal deferred final examination process. Such an application must be made to the student's Faculty office within 48 hours of the missed examination and must be supported by a Statutory Declaration or other appropriate documentation. (University of Alberta Calendar 23.5.6.) Deferred examinations are a privilege and not a right; there is no guarantee that a deferred examination will be granted. Misrepresentation of Facts to gain a deferred examination is a serious breach of the Code of Student Behaviour. If granted, the deferred final exam will be held from 9:00 to 12:00 (meeting 8:30 outside CAB 357) on Saturday, May 5, 2018.
    The University of Alberta is committed to the highest standards of academic integrity and honesty. Students are expected to be familiar with these standards regarding academic honesty and to uphold the policies of the University in this respect. Students are particularly urged to familiarize themselves with the provisions of the Code of Student Behavior and avoid any behaviour which could potentially result in suspicions of cheating, plagiarism, misrepresentation of facts and/or participation in an offence. Academic dishonesty is a serious offence and can result in suspension or expulsion from the University.
    All forms of dishonesty are unacceptable at the University. Any offense will be reported to the Associate Dean of Science who will determine the disciplinary action to be taken. Cheating, plagiarism and misrepresentation of facts are serious offenses. Anyone who engages in these practices will receive at minimum a grade of zero for the exam or paper in question and no opportunity will be given to replace the grade or redistribute the weights. As well, in the Faculty of Science the sanction for cheating on any examination will include a disciplinary failing grade (no exceptions) and senior students should expect a period of suspension or expulsion from the University of Alberta.
    Your student photo I.D. is required at exams to verify your identity. Students will not be allowed to begin an examination after it has been in progress for 30 minutes. Students must remain in the exam room until at least 30 minutes has elapsed. Electronic equipment cannot be brought into examination rooms.
    Cell phones are to be turned off during lectures, labs and seminars; they are not to be brought to exams. Audio or video recording, digital or otherwise, of lectures, labs, seminars or any other teaching environment by students is allowed only with the prior written consent of the instructor or as a part of an approved accommodation plan. Student or instructor content, digital or otherwis, created and/or used within the context of the course is to be used solely for personal study, and is not to be used or distributed for any other purpose without prior written consent from the content author(s).
    Eligible students have both rights and responsibilities with regard to accessibility-related accommodations. Consequently, scheduling exam accommodations in accordance with SAS deadlines and procedures is essential. Please note adherence to procedures and deadlines is required for U of A to provide accommodations. Contact SAS ( for further information.
    Students who require additional help in developing strategies for better time management, study skills or examination skills should contact the Student Success Centre (2-300 Students' Union Building).
    Students who require additional help with assignments or have questions about the course material in general are encouraged to visit the Decima Robinson Support Centre (528 Central Academic Building). Graduate students will be available to provide one-on-one help. In order to get maximum help during each visit, students are asked to be specific about the problem with which they are seeking help. The Centre is open Monday to Friday, 9:00-15:00.
    Any typographical errors in this Course Outline are subject to change and will be announced in class.
    Titanic00 updated