Master of Science in Statistical Data Science
The purpose of the program is to deliver a comprehensive curriculum in the fields of statistics and data science to prepare students with diverse backgrounds (including statistics, mathematics, computer science, engineering, and other quantitative fields) for the data science workforce or a doctoral program. The program curriculum emphasizes the following aspects. First, students will be trained in-depth in modern statistical and machine learning techniques in addition to the classical statistics theory and applications. Second, they will learn and polish computational skills for various types of data sets, including large-scale data ubiquitous in business, technology, and science. Third, these two aspects of the program are built on a solid foundation of statistical theory and an understanding of mathematical principles behind techniques and algorithms; this blend is crucial for success in industrial and academic careers in the rapidly changing big data era. Finally, the program has built-in flexibility for different backgrounds and career plans through various electives.
Applicants to the program must:
- hold a baccalaureate degree from a regionally accredited institution, or shall have completed equivalent academic preparation as determined by the appropriate campus authority;
- be in good academic standing at the last college or university attended;
- have a 3.0 GPA in their earned undergraduate degree or in the last 60 semester (90 quarter) units completed, or have earned a post-baccalaureate degree;
- have a baccalaureate degree in a quantitative field, including but not limited to statistics, mathematics, computer science, physics, engineering, or relevant fields. Successful applicants are expected to have completed three semesters of calculus, linear algebra, and upper-division undergraduate courses in probability and statistics with a grade of B or better. However, an applicant who is deficient in probability theory and/or statistics may be admitted conditionally on passing MATH 440 Probability and Statistics I and/or MATH 441/741 Probability and Statistics II satisfactorily during the first calendar year of study;
- submit a TOEFL score (minimum 550/80) or IELTS score (minimum 7.0) obtained within the past two years if their undergraduate degree is from a country where the official language is not English.
Program Learning Outcomes
- Apply statistical knowledge and computational skills to formulate problems, plan data collection, and analyze data to provide insight.
- Build and assess statistical and machine learning models, and employ a variety of formal inference procedures.
- Use mathematics to understand the underlying structure of common models used in statistical and machine learning.
- Prepare data for use with a variety of statistical methods and models, and recognize how the quality of data and data collection affect conclusions.
- Communicate effectively to a variety of audiences using oral, written, and visual modes.
Statistical Data Science (M.S.) — 30 Units
Program (12 Units)
|MATH 742||Advanced Probability Models||3|
|MATH 748||Theory and Applications of Statistical and Machine Learning||3|
|MATH 760||Multivariate Statistical Methods||3|
|MATH 761||Computational Statistics||3|
Upper Division/Graduate Electives (15 Units)
No more than 9 units should be from undergraduate only courses. Per student’s interested specialization and upon Graduate Advisor’s approval, the student will be recommended to choose a set of electives in the following areas: Probability and Statistics Electives, Mathematics Electives, Computer Science Electives, Biology Electives.
Culminating Experience (3 Units)
Candidates for the MS in Statistical Data Science must complete a Culminating Experience. Three options are available. Further information for these options can be obtained at the department website http://math.sfsu.edu
|MATH 895||Data Science Internship||3|
& MATH 896
|Culminating Experience Examination|
and Exam Preparation
|MATH 898||Master's Thesis||3|