Our statistics program leads to a Master of Science degree, offering students the perspectives and skills necessary to work as a statistician or data analyst in various sectors of the economy (banking, defense, health care, pharmaceuticals, government, insurance, biotech, agriculture, and many more). Students also have all the tools to continue their statistics education and obtain a Ph.D. if desired. The department has ready access to state-of-the-art computing and library resources.
The program also has an internship program for interested students. This cooperative effort trains students by a combination of formal university graduate courses and hands-on application in an industrial setting. The objective of the internship is to introduce the student to the application of applied statistics as a complement to the theoretical foundations learned in the classroom. Delaware’s many chemical, pharmaceutical and industrial companies provide a unique opportunity for statistics interns because of the problem-solving nature of their work, and the availability of experienced statisticians to guide and mentor our students.
Begin assembling your required application materials as electronic documents before completing the online graduate application. Do not mail any documents. Applicants must submit all materials directly to the University Office of Graduate and Professional Education using the online admission process before admission can be considered.
The total credits required for the degree are 33. If the student lacks background knowledge for one or more courses, prerequisite courses may need to be taken that do not count toward the degree.
Graduate students learn how to analyze, interpret and assess the validity of logistic regression and generalized linear models, and various applied contexts such as medicine, marketing, risk management, and online learning. Professors introduce modern topics such as high-dimensional logistic regression with Lasso and logistic regression in nonparametric or semi-parametric settings (generalized additive model). In addition to binary or multi-categorical data, Poisson regression and Negative Binomial regression for count data analysis will be studied. The course will primarily use procedures in the SAS system to do data analysis. The course will also introduce R software packages for high-dimensional logistic regression and generalized additive models, two modern machine learning techniques.
This applied multivariate analysis and statistical machine learning course introduces a variety of statistical methods for multivariate analysis and machine learning, involving statistical computing mostly with R and Python. The course topics include:
Random vectors and random matrices,
Multivariate normal distribution,
MANOVA (Multivariate analysis of variance),
Principal component analysis (PCA),
Canonical correlation analysis (CCA),
Linear and Quadratic discriminant analysis (LDA and QDA),
Resampling methods including Cross-Validation (CV) and Bootstrap,
Regression and classification trees (CART),
Support Vector Machines (SVM),
Online recommendation system,
Deep neural network,
Partial least squares, and
Sufficient dimension reduction.
This applied time series analysis course covers important topics in time series analysis, including the Box and Jenkins techniques of fitting time series data, ARMA models, ARIMA models, seasonal models, ARCH models, GARCH models, transfer function models, vector autoregression models, forecasting, frequency domain methods, recurrent neural networks, long short-term memory networks, gaussian processes and (hidden) markov models (time permitting). Professors focus more on methodology and data analysis than theory, involving an introduction to appropriate statistical packages in R and SAS software.
This course presents students with the basics of managing and summarizing data using the SAS System. Professors emphasize preparing data for analysis and creating attractive, readable reports for data summaries. Additionally, students will build the foundations and strategies to support future development of their SAS programming skills.