M.A. in Statistics
To qualify for the M.A. degree, the student must successfully complete eight term courses (with a 2-semester residency requirement unless accepted for part-time study over additional semesters) with an average grade of HP or higher and receive at least one grade of honors. Courses are chosen in consultation with the Director of Graduate Studies (DGS) at the start of each semester. Course selections become official only with the DGS’s signature. Courses in other departments can also be taken with permission from the DGS.
Specific requirements: It is hard to describe a typical course of study, because students come to the M.A. program with widely differing backgrounds. The minimal requirement is:
- all students must become acquainted with probability theory, at least at the level of Stat 538 or Stat 541.
- all students must learn some statistical theory, at least at the level of Stat 542.
- all students must gain some experience at working with real data, at least at the level of Stat 530, but preferably at the level of Stat 661 or higher.
Some students may be able to take higher level courses, such as STAT 610 (Statistical Inference) or STAT 625 (Statistical Case Studies).
Here, we provide some examples of courses taken by recent M.A. students.
In the past, some M.A. graduates have applied for Special student status (changed to “visiting student” as of 2022) in order to take additional graduate courses at Yale. Students petition for this status during the 2nd semester of their M.A. program and are required to complete the degree and graduate on time before beginning their study with visiting student status. Visa requirements may require full-time study (if applicable). The application is available here, but students should discuss with the DGS before applying.
M.S. in Statistics and Data Science
To qualify for the M.S. degree in Statistics & Data Science, the student must successfully complete twelve term courses (with a 3-semester residency requirement unless accepted for part-time study over additional semesters) with an average grade of HP or higher and at least two grades of H. Courses are chosen in consultation with the Director of Graduate Studies (DGS) at the start of each semester. Course selections become official only with the DGS’s approval. Elective courses in other departments are encouraged and are taken with permission from the DGS.
Specific requirements: It is hard to describe a typical course of study, because students come to the M.S. program with widely differing backgrounds and professional objectives. The minimal requirement with seven courses establishing a breadth of training is:
- all students must become acquainted with probability theory, at least at the level of S&DS 538 (Probability and Statistics) or S&DS 541 (Probability Theory).
- all students must learn some statistical theory, at least at the level of S&DS 542 (Theory of Statistics) or S&DS 612 (Linear Models).
- all students must gain some experience at working with real data in S&DS 625 (Case Studies).
- all students must take at least two courses in methods of data science, from a list of courses including (but not necessarily limited to): S&DS 563 (Multivariate Statistics), S&DS 565 (Previous: Applied Data Mining and Machine Learning; Current: Introductory Machine Learning), S&DS 661 (Data Analysis), S&DS 665 (Intermediate Machine Learning), CPSC 663 (Deep Learning Theory and Applications), S&DS 630 (Optimization Techniques), S&DS 631 (Computation and Optimization), S&DS 632 (Advanced Optimization Techniques), S&DS 668 (Nonparametric Estimation and Machine Learning), S&DS 669 (Statistical Learning Theory).
- all students must take at least two courses relating to efficient computation and Big Data. This could include (but is not limited to): S&DS 562 (Computational Tools for Data Science), S&DS 662 (Statistical Computing), BIS 620 (Data Science Software Systems), BIS 557 (Computational Statistics), BIS 634 (Computational Methods for Informatics), CPSC 524 (Parallel Programming Techniques), CPSC 526 (Building Distributed Systems), CPSC 527 (Object-Oriented Programming), CPSC 537 (Database Systems), CPSC 640 (Topics in Numerical Computation), or as approved by the DGS.
In addition, students are encouraged to complete an applied practical project (S&DS 626 or satisfied via an appropriate summer internship with S&DS 695) or suitable independent study. This would count for course credit, but is not a guaranteed opportunity; the student must identify suitable internships or projects and receive departmental approval. Credit for internships is only possible during the summers, and students studying on visas should consult with OISS.
The remaining four or five elective courses are selected in consultation with the DGS. Although some of these are likely to add depth to the areas outlines above, they may include coursework in specific areas of application including, but not limited to: Engineering and Applied Science, Economics, Computer Science, Linguistic, and Biostatistics. S&DS 627/628 (Statistical Consulting when taken for a full year) may also count as one of these electives.
Students typically complete the degree requirements during 3 consecutive semesters of study. Visa requirements (if applicable) may require full-time study. Some students may complete an internship during their 3rd semester and finish their studies during the 4th semester. Other students may, by petition during their 3rd semester and if not prohibited by visa requirements, complete their final coursework part-time during the second year. Students who finish at the end of their 3rd semester may request “visiting student” status for additional coursework in a 4th semester. The application is available here, but students should discuss with the DGS before applying.