Statistics

PhD Program

Courses

The following requirements apply to all current G1 students and those who enter the program during or after the fall of 2018.

The formal residence requirement for the PhD is sixteen four-credit courses devoted to advanced study.

Of the sixteen four-credit courses, students must take Statistics 210, 211, 212, 213, 220, 230 and 244; students must also take 300, 303hf, 314, 366hfra, 366hfrb, and at least two other 300-level special-topics courses.

Within the Statistics Department all courses marked 200 and above in the course catalog qualifies as advanced study, as does TIME(R), while Statistics Department courses marked 199 and below do not qualify.

Advising

As almost all PhD students are involved in research from the beginning of their program, entering PhD students are assigned faculty advisors based on mutual research interests whenever possible. New students’ research interests are determined by information provided by applicants in their Statements of Purpose. Students familiar with the department are given the option of requesting a particular faculty advisor. Some students stay with the initial advisor throughout their program, including the writing of the PhD dissertation, while others opt to change advisors as their interests change or evolve.

All PhD students are encouraged to consult with the director of graduate studies or department chair, when desired, in addition to their assigned advisor.

Students writing dissertations have one or more primary advisor(s) and two additional readers assigned by the faculty or suggested by the student and approved by the faculty. The primary advisor is typically the same faculty member providing the student’s research support.

In the case of conflicts between advisor and student, the department chair should be consulted first.

Qualifying Examination

All PhD students must pass a written qualifying examination in statistics. Students normally take the examination in January of their second year. The examination has two parts: 1) statistical methodology, including probability, and 2) applied statistics, including statistical design and data analysis.

The two parts of the examination are graded separately. A student may receive an unconditional passing grade on one or both parts. A student who fails one or both parts of the examination must retake the examination the next time it is given. Students who fail twice must withdraw from the program.

Post-Qualifying Talks (Research Presentations)

All students are required to attend Statistics 300: Research in Statistics. Students in their third year and above are required to present each year in the Statistics 300 course. The presentations, made to department faculty and students, are brief summaries of the student’s research and progress on qualifying papers or dissertation. Students in the first and second years are encouraged to volunteer to give a talk if this is helpful for their research.

Qualifying Paper

The qualifying paper is intended to provide the student with an opportunity to explore a serious topic in statistics and to express the findings in a written document. The work need not be original, but it should demonstrate an independent understanding of the topic, knowledge of the tools of research, and clarity of exposition. The effort involved is expected to require no more than the equivalent of one term at one-third time. The paper should be submitted and accepted by the department as early as possible, during the year following the qualifying examination. Prior written permission from the director(s) of graduate studies is required for a delay in the qualifying paper submission.

Dissertation

Each student is expected to exercise initiative in seeking out a dissertation topic, a faculty advisor who will take primary responsibility for supervising the student’s work, and two additional readers. The PhD dissertation is expected to be a research contribution of high quality, adding to the knowledge of either the theory or practice of statistics. A PhD dissertation in statistics may also consist primarily of an innovative analysis of a specific, complex body of data in some substantive field. Generally, the material in a PhD dissertation should be publishable in a referred journal.

In the fall term of the G3 year (after passing the qualifying exams) students must notify the department of their dissertation advisors. Students are required to update the department, at least once each semester, of their dissertation advisors. This would include their primary advisor plus any additional faculty who will advise or collaborate on the dissertation. Students who encounter difficulty in identifying appropriate advisor/s should consult with their primary advisor(s), the director of graduate studies, or the department chair. Students who wish to use as an advisor someone who is not a member of the department can and should request the approval of the department faculty. Also in the fall term of the G3 year, all students must submit to the faculty a preliminary title, one-page summary of the proposed topic, with names and signatures of at least two dissertation advisors.

Dissertation Defense

One copy of the completed dissertation must be submitted for consideration by the department faculty at least four weeks prior to the oral dissertation defense. The faculty will consider the submitted dissertation and make recommendations, which typically lead to revisions. Students will defend their dissertation by making a presentation in Stat 300 and in a closed presentation to their committee and other faculty members. After the defense, the faculty, with the explicit advice of three or more faculty readers nominated by the department, vote on the completed dissertation as submitted in finished form, which must conform to the requirements described in The Form of the PhD Dissertation. The approved final dissertation can then be submitted to the registrar. The time from the defense to the final vote is ordinarily about two weeks. A prospective sixth-year or more advanced student must submit evidence of significant dissertation progress to a dissertation advisor or committee each year. This evidence of progress may, at the department’s discretion, take the form of a dissertation chapter completed, manuscripts submitted for publication, abstracts of papers delivered at professional meetings, or other evidence as specified by the dissertation director.

Teaching Fellowships and Research Assistantships

All PhD candidates are normally required to teach and/or to work as research assistants beginning in their second year. The current Statistics Department teaching expectation is as follows: The Department considers teaching experience to be an important educational component of the PhD program. Most students will complete that component by leading four sections over the course of their PhD careers, covering at least two different courses. Many students elect to do additional teaching for a variety of reasons, including funding living expenses, addressing the needs of the Department, and/or obtaining additional professional experience.  The Department’s aim is to not ask students to teach more than eight sections in total. Research funding considerations may affect these normal teaching and research workloads. When research funds permit, the department aims to assign final-year students to lighter-than-normal teaching loads to enable them to make progress on their dissertations.

Master of Arts (AM) Program

The Department of Statistics awards terminal AM degrees, as well as AM degrees to students who are continuing in the PhD program. The department will consider for the AM degree PhD candidates in other fields at Harvard for whom a statistics minor is appropriate, well-prepared undergraduates eligible for the AB/AM program, and candidates with appropriate mathematics backgrounds who demonstrate motivation for pursuing a terminal AM degree.

Students in the AM program are advised by the director(s) of graduate studies for the master's program and department faculty.

  • Minimum of one year of full-time residence. Requests for part-time study in the first year must meet the GSAS special circumstances criteria.
  • Requires the satisfactory completion of eight four-credit courses taken within the Department of Statistics and approved by the student’s department advisor, at the level of Statistics 110 and above, with at least five 200-level course from the list below. The actual course of study will vary according to the student’s interest and preparation and will be determined in consultation with the student’s advisor.
  • Statistics 210 and Statistics 211 are required.
  • The eight statistics courses must also include two courses that are at the interface of theory and application. Examples of such courses are Statistics 115, 120, 121, 131/231, 139, 140, 149, 160/260, 170, 183, 186, 220, 221, 225, 230, 232r, 240, 244, and 245. The department will maintain an up-to-date list of courses that meet these criteria.
  • AM students must earn at least a B average in Statistics courses; no more than one C may count toward the eight courses for the AM degree. Students must also meet the GSAS grade requirements..
  • CS 181 and/or  CS 281 can count towards the 8 Statistics courses but not toward the 5 200-level courses.
  • Terminal AM students must earn at least a B average in their Statistics courses and no more than one C.
  • Terminal AM students may take 300-level courses, with the permission of the instructor. However, only one 300-level course will count toward the required 8 letter-graded half-courses. 
  • No thesis, general examination, qualifying paper, or cognate field is required for the AM degree.