Skip to content Skip to navigation

MS in Statistics: Data Science

The increasing importance of big data in engineering and the applied sciences motivates the Department of Statistics and ICME (Institute for Computational and Mathematical Engineering) to collaboratively offer a M.S. track that trains students in data science with a computational focus.

This focused M.S. track is developed within the structure of the current M.S. in Statistics and the M.S. program in ICME. Students in the program will develop strong mathematical, statistical, computational and programming skills through the M.S. requirements, and they will gain a fundamental data science education by focusing 18 units of elective courses in the area of data science and related courses. Upon the successful completion of the Data Science M.S. degree students will be prepared to continue on to their Ph.D. in Statistics, ICME, MS&E, or Computer Science or as a data science professional in industry. Completing the M.S. degree gives no guarantee or preference for admission to the Ph.D. program.

The track will attract both engineering or science students interested in better understanding the mathematical and statistical underpinnings of data science, as well as mathematically oriented students who are looking to gain expertise in data science and applications.

The M.S. in Data Science track is overseen by a steering committee comprised of ICME and Statistics faculty members.

The Data Science track is not an online degree program.

We do not accept applications for internal/coterminal students.

Students interested in applying to the Data Science program must apply externally. Please see our admissions pages for more information: https://statistics.stanford.edu/academics/ms-ext-apply.

Application and Selection Process

For those who want to pursue the Data Science subplan, you must apply to the Statistics Master's degree program and then declare your preference by selecting the subplan in the "Department Specialization" option.
Applicants who select Data Science will only be considered for the Data Science program, not in addition to the Statistics M.S. degree program.

Admission to the M.S. program is made by the Statistics admissions committee, which will have representation from the Data Science track steering committee.

It is the applicant's responsibility to ensure he or she meets all eligibility requirements before applying.

Students should have successfully taken an equivalent of Linear Algebra and one other statistics courses to be eligible.

Coursework

The coursework follows the requirements of the traditional ICME M.S. degree with additional restrictions placed on the general and focused electives. As defined in the general Graduate Student Requirements, students have to maintain a grade point average (GPA) of 3.0 or better and classes must be taken at the 200 level or higher. Students satisfying the course requirement of the Data Science track do not have to satisfy the other course requirements for the M.S. in Statistics

The total number of units in the degree is 45, 36 of which must be taken for a letter grade.

Submission of approved Master's Program Proposal, signed by the master's adviser, to the student services specialist by the end of the first quarter of the master's degree program. A revised program proposal is required to be filed whenever there are changes to a student's previously approved program proposal.

 Data Science Program Proposal Form (PDF)

Students must demonstrate breadth of knowledge in the field by completing five core areas as required by ICME.

Requirement 1 - Mathematical Core (12 units)

These 12 units must be taken for a letter grade.

The recommended Mathematics core courses for the Data Science track are:

  • CME302, Linear Algebra
  • CME304, Numerical Optimization or CME364A,  Convex Optimization
  • CME305, Discrete Mathematics

In addition to these three core courses, the students are required to take a course in stochastics. They can take either CME308 or an equivalent course approved by the steering committee.

Requirement 2 - Advanced Scientific Programming and High Performance Computing Core (6 units)

To ensure that students have a strong foundation in programming all students will be required to take 6 units of advanced programming, with at least 3 units in parallel computing. These 6 units must be taken for a letter grade.

Approved courses include:

  • CME212 Advanced Programming for Scientists and Engineering
  • CME214 Software Design in Modern Fortran for Scientists and Engineering
  • CS107, Computer Organization and Systems
  • CS249B, Large Scale Software Development

And for parallel/HPC (at least 3 units required):

  • CME213 Introduction to Parallel Computing using MPI, openMP and CUDA
  • CME342, Parallel Methods in Numerical Analysis
  • CS149, Parallel Computing
  • CS315A, Parallel Computer Architecture and Programming
  • CS315B, Parallel Computing Research Project
  • Also CS316, or CS344C

Similar courses may be approved by the steering committee.

Students who do not start the program with a strong computational and/or programming background will take an extra 3 units to prepare themselves by, for example, taking CME211 Programming in C/C++ for Scientists and Engineers or an equivalent course, such as CS106A/B or CS106X.

Requirement 3 - Statistics Core (12 units)

The following 12 units must be taken for a letter grade.

The curriculum for the Data Science track requires 12 units of focused coursework in Statistics consisting of the following courses:

  • STATS200, Introduction to Statistical Inference
  • STATS203/305, Regression Models / Statistical Modeling
  • STATS315A, Modern Applied Statistics: Learning
  • STATS315B, Modern Applied Statistics: Data Mining

or equivalent courses as approved by the steering committee.

Requirement 4 - Domain Specialization or preparatory courses (9 units)

Three courses in specialized areas. One or two of these courses may be used by the students that enter the program with insufficient linear algebra or programming experience to prepare for the core requirements in the M.S. track.

Of the following 15 units in Requirements Four and Five combined, 6 units must be taken for a letter grade. Specialized courses include courses that further deepen the data science core. Some possibilities include:

  • CS347, Parallel and Distributed Data Management
  • STATS290
  • CS448, Topics in Computer Graphics
  • CS224W, Social and Information Network Analysis
  • STATS366/BIOS221, Modern Statistics for Modern Biology, Holmes/Martin (Summer)
  • Psych204A, Human NeuroImaging Methods, Wandell/Dougherty (Autumn)
  • Psych303, Human and Machine Learning (not given this year)
  • OIT367, Analytics from Big Data, Bayati (Winter)
  • BioMedin215, Data Driven Medicine, Shah (Autumn)
  • Energy240, Geostatistics, tbd (Spring)
  • BIOE214, Representations and Algorithms for Computational Molecular Biology, Altman (Autumn)

Requirement 5 - Practical component (6 units)

The students need 6 units of practical component that may include any combination of:

  • A capstone project, supervised by a faculty member and approved by the steering committee: the capstone project should be computational in nature; students should submit a one-page proposal, supported by the faculty member, to the steering committee (gwalther@stanford.edu) for approval.
  • Clinics, such as the new Data Science Clinic offered by ICME starting Fall 2013.
  • Other courses that have a strong hands-on and practical component, such as STATS390 (Statistical Consulting).

Data Science Sample Schedules

The Data Science track schedule typically spans 5 quarters.

5 quarter schedule for most students:
Year 1:
Aut: CME 200, CME211, STATS200
Wtr: CME212, CME364A, STATS203
Spr: STATS315B, CME308, domain Spec.
Year 2:
Aut: CME302, HPC course (or take CME213 in spring), practical
Wtr: CME305, STATS315A, practical

 

5 quarter schedule for students who are well prepared:
Must have taken the equivalent of CME200 and STATS200 before starting the program.
Year 1:
Aut:  CME211, STATS 305, domain spec
Wtr: CME212, CME364A, STATS315A
Spr: CME213, STATS315B, CME308
Year 2:
Aut: CME302, STATS263, practical
Wtr: CME305, domain spec, practical

 

4 quarter schedule:
This schedule is very demanding and students typically prefer the experience gained with 
a 5 quarter schedule. Student must have taken the equivalent of CME200 and STATS200 
before starting the program.
Year 1:
Aut: CME211, STATS305, domain spec
Wtr: CME212, CME305, CME364A, STATS315A
Spr: CME213, STATS315B, CME308, practical
Year2:
Aut: CME302, STATS263, domain spec, practical
 

Notes:
1. CME200 and CME211 count as Domain Specific courses.
2. CME302 requires the equivalent of CME200 as prerequisite.
3. STATS305 requires the equivalent of STATS200 as prerequisite.
4. STATS315A requires the equivalent of STATS200 and (STATS203 or 305) as prerequisite.
5. STATS200 and CME200 can be completed over the summer; STATS200 taken in the summer prior to starting the program can be counted towards the master’s program requirements. However, CME200 is not available for taking for credit over the summer quarter; only the lectures are available for review online and hence, completion of CME200 review over the summer  cannot be counted towards the degree requirements.