Department of Statistics offers M.S. and Ph.D. degrees in Statistics.
M.S. degree in Statistics:
- Requires at least seven courses with credit
- The curriculum consists of compulsory courses, departmental elective courses, and non-departmental elective courses.
- Compulsory courses are theoretical statistics courses and aim to have students acquire a sound understanding of the theoretical basis of statistics.
- Departmental elective courses aim to have students get skilled in modeling and computational methods for statistics applied in various different areas such as economics, finance, marketing, sociology, psychology, genetics, and medicine.
- The degree requires conducting a research and writing a thesis. A thesis is usually one of the following: methodological, computational, empirical, or theoretical. In a typical methodological M.S. thesis, student compares and evaluates different available statistical models and estimation methods through a simulation study. In a computational M.S. thesis, computational statistical methods are the emphasis. In an empirical M.S. thesis, a novel data set is the central to the thesis and a collection of appropriate existing statistical methods are employed for the analysis.
Ph.D. degree in Statistics:
- Our Ph.D. program is structured with the objective of preparing students for academic careers and for industrial and government positions that involve consulting and research in new statistical methods.
- Requires at least ten courses with credit
- The curriculum consists of compulsory courses and other departmental courses
- Compulsory courses aim to have students absorb the probability theory and the theory of statistical inference.
- Other courses towards this degree are designed to enable student to get an in-depth understanding of the methodologies related with different types of data such as data obtained from designed experiments and multivariate data in a broader sense.
- The degree requires conducting a research and writing a thesis. A qualified Ph.D. thesis is based on a research on a previously unknown area. The research may be motivated by a novel or an interesting existent data set for which one lacks a computationally efficient way of modeling that provides estimators with desired statistical properties at the same time.