The Online Master of Science in Business Analytics

In the area of probability, this course covers the concepts of discrete and continuous probability distributions as well as conditional probability. It also covers basic statistics, which can be thought of as a set of tools for interpreting data. These include descriptive statistics, which permit us to describe basic characteristics of data, including the computation of means, standard deviations and ranges of a data set. This course also covers inferential statistics, which are methods for uncovering deeper insights from the data, such as hypothesis testing. Finally, the course considers data visualization as an integral part of data analysis.

This course provides a set of programming skills using the R programming language, which is widely used in business analytics for statistical computations.

This course provides a foundation of Python programming skills for business analytics including knowledge of Python data types, facilitating repeated execution through the application of loops, using conditional statements, programming the input and output of data, the use of Python packages, and the construction of functions.

This course provides a set of linear algebra tools for performing business analytics including vector-matrix multiplication, Gaussian elimination, computing determinants, computing matrix rank, computing matrix column and row spaces, performing eigenanalysis, and performing principal components analysis.

An interactive virtual seminar that helps set students up for success in this rigorous program. Led by your Student Success Coordinator, your OMSBA Orientation will cover an array of topics, including:

• An introduction to William & Mary and the Raymond A. Mason School of Business
• Tips for managing your work/life balance while in grad school
• Educational expectations for the program
• Academic policies and standards
• The virtual classroom and learning management system (LMS)
• Tips for networking while studying online
• Advice for making the most of your digital learning experience

Completion of this course ensures that students have sufficient skills in Excel. This non-credit course will be available in Canvas throughout the program for students' reference as needed.

This course will include a survey of the state-of-the-art in business analytics, examining companies that have used business analytics for competitive advantage and how they have done so. This course will teach business acumen and how the field of analytics fits within the context of business. Topics will include business metrics as used for performance measurement and incentives, communicating with impact, visualization, and the functions of a company—how they interact, what data they have, and their development and deployment of algorithms. The course will also include a survey of opportunities for problem solving using business analytics in operations, supply chain, human resources, finance and marketing, as well as an introduction to the tools that are covered in the remainder of this program.

Probability and Statistics is a foundation course in the study of business analytics. It provides an understanding of the principles associated with modeling of stochastic processes. The topics will include probability theory—important probability distributions, sampling from distributions and the interaction of multiple stochastic processes; regression; statistical analysis—descriptive/inferential/predictive statistics, multivariate statistics and time series models; and modeling—modeling concepts, Monte Carlo simulation and decision analytics. Students will also be introduced to a variety of statistical modeling packages.

Organizations store data in two types of databases: operational and analytical. Operational database topics include database requirements, entity relationship modeling, relational modeling database constraints, update anomalies, normalization, Structured Query Language (SQL) and data quality. Analytical database topics include data warehousing concepts, dimensional modeling (star schemas), data warehouse/data mart modeling approaches, the extraction/transformation/load (ETL) process, online analytical processing (OLAP)/business intelligence (BI) functionalities and the data warehouse/data mart front end. Once data is cleaned and stored, data visualization is used to most effectively communicate information contained in the data. The course covers data visualization principles drawn from the fields of statistics, perception, graphic and information design, and data mining. Students will learn visual representation techniques that increase the understanding of complex data and models. Topics include charts, tables, graphics, effective presentations and dashboard design.

This course is designed to provide students with a deep understanding of the theory and practice of regression and classification, two of the most commonly used techniques in the data scientist's toolkit. These predictive analytics techniques are important members of a family of analytics often referred to as machine learning techniques. The programming language R is used extensively in labs and assignments in this class and subsequently reappears in other classes throughout the program.

This course is designed to provide students with a deep understanding of machine learning and big data, including more elaborate techniques that extend the coverage from Machine Learning I. The data storage and retrieval techniques that have served the information processing industry for decades have proven inadequate in the face of the huge collections of data presently being created by the internet and the so-called "Internet of Things." Businesses today require a new set of technologies that are specifically designed to deal with these huge data sets. In this course, the Hadoop environment and Amazon Web Services (AWS) will be used to process large-scale data sets.

Optimization is an analytics methodology designed to yield the best solution to a given problem. Students are exposed to theory and applications of optimization including linear programming, non-linear programming, discrete optimization and specialized networks. This course includes discussion about the difficulties of accurately representing real-world processes with a mathematical model. Most business problems are too large or too complex to be solved optimally, where the strict definition of "optimal" means finding the provably best solution. Finding a solution that approximates the optimal solution is, therefore, the predominant mode of problem solving found in industry: these are called heuristic solutions. Many companies gain a competitive advantage by constructing heuristics that either find better solutions than do their competitors or find solutions more quickly. This course focuses on achieving such results by programming custom algorithms, which are a sequence of steps taken to provide a solution to a problem.

The theme of this course is "natural models and artificial intelligence." The course considers natural models of intelligence and their artificial equivalents. It shows how viewing natural intelligence is an effective mindset and it describes the key analytics tools required for designing and executing some business processes competently. A majority of the course is devoted to the topic of neural networks, although other methods are included, such as genetic algorithms, simulated annealing and swarm intelligence.

This experiential practicum course includes a comprehensive business analytics project that the student will complete from start to finish, integrating the skills that have been acquired from their previous coursework in the business analytics program. They will define and frame a complex problem, develop a systematic approach to solving it using analytics, identify methodologies that are suited to the problem, quickly prototype solutions with those methodologies to identify the best approach, and, ultimately, generate an innovative solution and persuasively convey that solution using data visualization techniques and communication skills.