Statistics

2014-15 Tuition

Research degree: $20,800; M.P.S.: $47,050

Application deadlines

Ph.D.: Fall, Jan. 1; no spring admission
M.P.S.: Fall, March 15; Spring, Nov. 15 (Note: Only limited number of Spring term admissions are available. Applicants for Spring admission may also be a considered for the subsequent Fall term upon request.)

Requirements summary

  • all Graduate School Requirements, including the TOEFL Exam for Non-Native English Applicants
  • two letters of recommendation required, but four allowed
  • GRE general test for M.P.S., and Ph.D.

Degrees

  • M.P.S.(A.S.)
  • Ph.D.

Subjects

  • Applied Statistics (M.P.S.(A.S.))
  • Statistics (Ph.D.)

Major concentrations

  • applied statistics
  • biometry
  • decision theory
  • econometrics
  • engineering statistics
  • experimental design
  • mathematical statistics
  • probability
  • sampling
  • social statistics
  • statistical computing
  • stochastic processes

The Field of Statistics offers two graduate degree programs: an MS/PhD degree in Statistics and a Masters of Professional Studies degree in Applied Statistics. The field does not offer a Masters degree in Statistics.


The PhD program is intended to prepare students for a career in research and teaching at the University level or in equivalent positions in industry or government. A PhD degree requires writing and defending a dissertation. Students graduate this program with a broad set of skills, from the ability to interact collaboratively with researchers in applied fields, through the formulation and computational implementation of novel statistical models and methods to demonstrating mathematically that these methods have desirable statistical properties. Cornell's PhD alumni have gone on to high profile positions in all of academia, industry and government.

The Master of Professional Studies (M.P.S.) degree in Applied Statistics is for persons interested in professional careers in business, industry or government. The M.P.S. program has three main components:

  •  A two-semester core course covering a wide range of statistical  applications, computing, and consulting
  • An in-depth statistical analysis project
  • Elective coursework drawn from the resources of the Department of Statistical Science.

The program can be completed in one year by a well-prepared student with the equivalent of an undergraduate degree in statistics or applied mathematics. Students with less preparation can make up any missing prerequisites while at Cornell; in this case the program will take one to two years to complete.

M.P.S. or M.S./Ph.D.?
Statistics does not offer admission for those interested a terminal master's degree, but we do offer admission for those interested in pursuing a master's leading to a Ph.D. We also offer the M.P.S. in Applied Statistics, which is normally a one-year program that does not carry financial aid.

The M.P.S. is intended for persons who want a short-term (one year) master's degree so as to go into business, industry, or government statistical work. The M.P.S. is not equivalent to an M.S. on several counts: the M.P.S. has a project (a large-scale data-analysis project) rather than a thesis or a qualifying exam (which would be the case for an M.S.). The mathematical probability/statistics component of the M.P.S. is less than it would be for an M.S. (which would be considered the first part of a Ph.D.).

The admissions procedures are completely independent: at Cornell, if you want to go on for a Ph.D. after the M.P.S. you must to apply as a new student to the Ph.D. program; you would be considered as part of the "pool" of Ph.D. applicants and, if admitted, you might be able to apply some of your M.P.S. coursework, but there is no guarantee. The Ph.D. in Statistics at Cornell enrolls about 2 to 4 students each year; the M.P.S., about 20 to 25.

If you are applying for the M.P.S., please make clear your clear if you are applying for Option 1 or Option 2.

Application:
In their transcripts applicants must show strength in the mathematical sciences. Applicants must also demonstrate strong motivation for advanced study in statistics. Submission of GRE general test scores is recommended.

John Abowd -- Concentrations: sampling; applied statistics; econometrics; Research interests: data confidentiality; record linkage; econometric analysis of linked data; analysis of labor markets
Jacob Bien -- Concentrations: mathematical statistics; applied statistics; Research interests:
James Booth -- Concentrations: biometry; experimental design; mathematical statistics; statistical computing; applied statistics; Research interests: computer intensive methods; generalized linear models; Monte Carlo simulation; statistical genomics
Florentina Bunea -- Concentrations: mathematical statistics; decision theory; engineering statistics; probability; Research interests: high dimensional modeling; sparsity; model selection; model averaging; non-parametric statistics; machine learning
John Bunge -- Concentrations: sampling; probability; applied statistics; Research interests: point processes; semi-Markov processes; species problems
Jim Dai -- Concentrations: probability; applied statistics; Research interests: Stochastic processing networks Fluid and di usion models of queueing networks Impulse, singular and drift controls of di usions Customer contact center management Patient flow management in hospitals Semiconductor wafer manufacturing Revenue management  Algorithm trading, orderbook dynamics
Thomas Diciccio -- Concentrations: mathematical statistics; applied statistics; econometrics; Research interests: likelihood inference; resampling methods; asymptotic approximations; linear models
Daniel Fink (Minor Member) -- Concentrations: statistical computing; applied statistics; Research interests: Statistics, Machine Learning, Spatial Statistics, Exploratory Analysis, Semiparametric Regression, Predictive Analytics, Data Analysis, Observational data, Crowdsourced Data, and Citizen Science
Peter Frazier -- Concentrations: probability; applied statistics; Research interests: Optimal learning, sequential decision-making under uncertainty, and machine learning, focusing on applications in simulation optimization, design of experiments, materials science, e-commerce and medicine.
Yongmiao Hong -- Concentrations: mathematical statistics; econometrics; Research interests: nonparametric testing; econometrics
Giles Hooker -- Concentrations: biometry; experimental design; mathematical statistics; statistical computing; stochastic processes; sampling; applied statistics; Research interests: statistical learning theory; nonlinear functional data analysis; diagnostic and graphical statistics; nonlinear regression analysis
J.T. Gene Hwang -- Concentrations: mathematical statistics; engineering statistics; applied statistics; Research interests: Genomics statistics; shrinkage procedures including estimation, confidence intervals and multiple tests
Thorsten Joachims -- Concentrations: mathematical statistics; statistical computing; engineering statistics; Research interests: machine learning; text-mining; statistical learning theory; information access
Alon Keinan -- Concentrations: biometry; statistical computing; applied statistics; Research interests: human population genetics; demographic inference
Nicholas Kiefer -- Concentrations: decision theory; applied statistics; Research interests: econometrics; risk management; Bayesian statistics
Thomas Loredo (Minor Member) -- Concentrations: statistical computing; applied statistics; Research interests: poisson processes;marked point processes; gaussian processes; astrostatistics; astroinformatics; bayesian statistics; bayesian computation; bayesian experimental deisgn;functional data analysis; statistical software development
David Matteson -- Concentrations: biometry; social statistics; mathematical statistics; statistical computing; stochastic processes; engineering statistics; econometrics; Research interests: financial econmetrics; non parametric statistics; spatio-temporal statistics; biostatistics; machine learning
Jason Mezey -- Concentrations: biometry; statistical computing; applied statistics; Research interests: quantitative genetics/genomics; statistical genetics; computational biology; pathway modeling; molecular evolution
Francesca Molinari -- Concentrations: mathematical statistics; sampling; econometrics; Research interests: econometrics; identification; survey methodology
Michael Nussbaum -- Concentrations: mathematical statistics; Research interests: mathematical statistics
Sidney Resnick -- Concentrations: stochastic processes; engineering statistics; probability; Research interests: applied probability; extreme-value theory; data network analysis
David Ruppert -- Concentrations: biometry; mathematical statistics; statistical computing; engineering statistics; applied statistics; Research interests: semiparametric regression; functional data analysis; splines and nonparametric estimation; astrostatistics; calibration and uncertainty analysis; environmental statistics
Gennady Samorodnitsky -- Concentrations: stochastic processes; engineering statistics; probability; Research interests: probability theory
Adam Siepel -- Concentrations: biometry; statistical computing; applied statistics; Research interests: comparative genomics; molecular evolution; population genetics; probabilistic graphical model
Patrick Sullivan -- Concentrations: biometry; experimental design; sampling; applied statistics; Research interests: Applied Statistics, experimental design, sampling
Bruce Turnbull -- Concentrations: biometry; experimental design; mathematical statistics; stochastic processes; engineering statistics; sampling; applied statistics; Research interests: biomedical statistics; reliability and life testing
Paul Velleman -- Concentrations: statistical computing; applied statistics; Research interests: statistical computing; robust exploratory methods
Marten Wegkamp -- Concentrations: mathematical statistics; Research interests: classification; copulas; empirical processes; high dimensional models; model selection; non parametric estimation; penalized empirical risk minimization
Martin Wells -- Concentrations: biometry; mathematical statistics; decision theory; sampling; applied statistics; econometrics; Research interests: Bayesian statistics; decision theory; empirical legal studies; epidemiology; social statistics; statistical bioinformatics; survival analysis
Dawn Woodard -- Concentrations: statistical computing; decision theory; engineering statistics; probability; Research interests: statistics

In the broadest possible terms, the purpose of the graduate program in the Field of Statistics is to prepare its students for a career in Statistical Science, with coverage that is sufficiently deep in its coverage of core principles and methods and sufficiently broad in its coverage of application areas to prepare its students for future employment and success in a diverse array of environments. Specific program goals include the following:

  • Attract the best possible graduate students and provide those admitted to the program with the financial and research resources necessary for achieving their educational goals.
  • Provide advanced training through coursework and mentored research to help graduate students gain the skills and experience needed to successfully pursue a career as a statistician in academia, government or industry.
  • Prepare graduate students for future leadership in research and scholarship, challenging those students to reach the highest possible level of achievement.
  • Facilitate completion of graduate degrees in a timely manner.

Assessment for M.P.S. and PhD degrees:

The M.P.S. program consists of coursework and an applied project and assessment is conducted through course exams, assignments and performance on the project.

The PhD program requires coursework, completion of an Admission to Candidacy Exam and a Thesis Defence. Depending on a coursework performance, students may also be required to sit a qualifying exam.

Upon completion of the Ph.D. degree, students will have

  1. Demonstrated mastery of statistical theory and methods;
  2. Achieved breadth and diversity of knowledge through elective coursework and research/teaching experiences;
  3. Demonstrated the ability to work collaboratively across disciplines, communicating statistical principles, methods and results to a lay audience;
  4. Demonstrated a high level of proficiency in oral and written communication skills appropriate for a career in either (i) advanced research and/or teaching at a college or university; or, (ii) advanced research in government and industry;
  5. Demonstrated the ability to independently conduct, document and defend original research contributions having the potential to advance the field of statistical science.

The M.P.S. program in Applied Statistics

The MPS program in Applied Statistics is a two-semester program designed to prepare students for careers as statisticians in industry.  Incoming students are expected to have mastery of calculus, linear algebra, and basic statistics as well as some computer programming experience.  Statisticians are in great demand by industry and nearly all of the program’s graduates obtain employment in their chosen field.  A few of them decide instead to pursue a PhD in statistics or another field.

Expected Learning Outcomes

Upon completion of the MPS degree, students will have

  1. Demonstrated mastery of basic statistical theory and methods;
  2. Developed proficiency in the use of statistical software;
  3. Achieved breadth and diversity of knowledge through elective courses;
  4. Demonstrated the ability to creatively use statistical methods to solve real-world problems.
  5. Demonstrated the ability to work in teams;
  6. Demonstrated a proficiency in oral and written communication skills appropriate for a career in industry.


Assessment of Learning Outcomes


The learning outcomes will be assessed by

  1. Completion of STSCI 5080 (Probability Models and Inference) or STSCI 4090 (Theory of Statistics) and STSCI 4030 (Linear Models and Matrices)
  2. Completion of STSCI 5010 (Applied Statistical Analysis), which teaches SAS, and STSCI 4030, which teaches R.
  3. Completion of a total of 30 credit hours of approved coursework.  Approved courses will be in statistics as well as courses with substantial statistics content in the areas of computer programming, bioinformatics, and financial engineering.
  4. Completion of the MPS project course, STSCI 5999. 
  5. Completion of the MPS project course, STSCI 5999.  In this course, teams of usually 4 students work under the direction of a faculty advisor and with a client in industry or (less commonly) in another department at Cornell.
  6. Completion of the required oral and written reports in STSCI 5999.
  7. Performance in these tasks will be assessed through grades in each of the classes listed above, and average elective grade for Outcome c.