Seminars in Computational Statistics
Courses in Computational Statistics
School of Computational Sciences
Department of Applied Statistics
School of Information Technology
Computational Statistics is an area of concentration in the PhD program in Computational Sciences that is administered by the School of Computational Sciences.
The PhD program requires
- 48 credit hours of course work beyond the baccalaureate degree
- 3 credit hours of seminars or colloquia
- successful performance on a comprehensive examination
- 24 credit hours of thesis research culminating in completion and public defense of a dissertation reporting on significant research in the field of computational statistics.
Because computational statistics is built on the mathematical theory and methods of statistics, the student must take a number of the usual graduate-level statistics courses, in addition to courses in computational statistics.
Dissertation Committee
After the student has become somewhat familiar with the program and has a preliminary idea of the area of research, the student selects a faculty member who is willing to direct the dissertation research and writing. The student then forms a dissertation committee with the consent of the dissertation director, the SCS graduate coordinator, and the SCS dean. The committee is must consist of a minimum of four members of which at least three must be SCS program faculty members. Non-GMU members may serve on the committee with the consent of the SCS program faculty members on the committee. The next major step is to agree on a program of study to be approved by the committee.Program of Study
The program of study includes the four CSI core courses:- CSI 700 Numerical Methods
- CSI 701 Foundations of Computational Sciences
- CSI 703 Scientific and Statistical Visualization
- CSI 710 Scientific Databases
The statistics courses in the program of study depend on what statistics courses have previously been taken, and on the specific area of computational statistics in which the student will work. These courses can be grouped as follows.
- The basic statistics courses (students with a Master's degree in statistics
would likely have had these already):
- STAT 544 Applied Probability
- STAT 554 Applied Statistics
- CSI 672 / STAT 652 Statistical Inference
- STAT 656 Regression Analysis
- Real analysis courses with statistical applications (students should take
at least two of these):
- CSI 778 Real Analysis and Statistics
- CSI 876 Measure and Linear Spaces
- CSI 877 Geometric Methods in Statistics
- A two-course sequence in mathematical statistics (students must take
both of these):
- CSI 972 Mathematical Statistics I
- CSI 973 Mathematical Statistics II
- Computational statistics courses (students should take
at least one of these):
- CSI 771 Computational Statistics
- CSI 773 Exploratory Data Analysis
- CSI 779 Topics in Computational Statistics
- CSI 979 Advanced Topics in Computational Statistics
- Other courses in general areas of
statistical theory and methods or in
specialized areas:
- STAT 655 Analysis of Variance
- STAT 657 Nonparametric Statistics
- CSI 678 / STAT 658 Time Series Analysis and Forecasting
- STAT 662 Multivariate Statistical Methods
- STAT 665 Categorical Data Analysis
- STAT 574 and STAT 674 Survey Sampling I and II
- STAT 673 Statistical Methods for Longitudinal Data Analysis
- CSI 842 Linear and Nonlinear Modeling in the Natural Sciences
- CSI 847 Wavelet Theory
- CSI 848 Mathematical Tomography
- CSI 865 Visualization and Modeling of Earth Systems and Space Sciences Data
- CSI 873 Statistical Methods in Astronomy
- CSI 903 Advanced Topics in Scientific Visualization
- CSI 904 Seminar in Scientific Visualization
- CSI 976 Statistical Inference for Stochastic Processes
- CSI 978 Statistical Analysis of Signals
In addition to the 48 hours of formal course work, three hours of colloquia and/or seminars are required. They can be selected from the following one-hour courses, which may be repeated for credit.
- CSI 898 Colloquium in Computational Statistics
- CSI 899 Colloquium in Computational Sciences
-
CSI 991 Seminar in Computational Statistics
A maximum of 24 credit hours of previous graduate course work may be applied toward the required 48 hours.
After completion of the 48 hours of course work, the next two steps are successful performance on a comprehensive exam and the approval by the student's committee of a dissertation topic and tentative outline.
The comprehensive exam normally consists of a written portion covering relevant theory, a computational component, and an oral examination. Samples of some past exam questions are available.
Following successful completion of the comprehensive exam and approval of the dissertation proposal, the student is admitted to candidacy.