The Computational Modeling and Data Analytics (CMDA) program draws on expertise from three primary departments at Virginia Tech with strengths in quantitative science: Mathematics, Statistics, and Computer Science. By combining elements of these disciplines in innovative, integrated courses that emphasize techniques at the forefront of applied computation, we teach a rich suite of quantitative skills for tackling today's massive data-based problems.
CMDA students learn to be dynamic problem solvers who can work collaboratively in interdisciplinary teams. Modern computational science draws on a variety of expertise. Our students get a great deal of experience practicing these skills, particularly in the CMDA Capstone Project course. Teams of CMDA majors spend the semester tackling an interesting data/modeling problem presented from a client (elsewhere on campus, or in industry). Examples of this year's projects include:
- Russian flu outbreak: a historical analysis of the morbidity and mortality of the 1889 epidemic
- Measuring the impact of Open Source Software on innovation
- Effectiveness of digital marketing campaigns for VT Athletics
- Gaining understanding of Medical Deserts on the DC metro area
The CMDA requirements give students the flexibility to dive deeply into a particular area through choice of electives, a disciplinary track, or a minor or double major. Possibilities include a CMDA-Physics track and a CMDA-Economics track.
Ultimately, our program is for anyone who seeks to better understand the world through data and computation.
What kind of jobs will CMDA majors qualify for?
- analytics for sports teams or internet/social media companies
- consulting in defense/space/homeland security
- modeling in the oil/gas/alternative energy sector
- data analysis for medical/pharmaceutical firms
- quantitative modeling in finance/insurance
- a host of other possibilities!
CMDA Course offerings
CMDA 2005: Integrated topics from quantitative sciences that prepare students for advanced computational modeling and data analytics courses. Topics include: probability and statistics, infinite series, multivariate calculus, linear algebra.
Prerequisites: MATH 1226 (C-) Co-requisite: MATH 2114
CMDA 2006: Intermediate linear algebra, regression, differential equations, and model validation;
Prerequisite: CMDA 2005, MATH 2114
CMDA 3605: Mathematical modeling with ordinary differential equations and difference equations. Numerical solution and analysis of ordinary differential equations and difference equations. Stochastic modeling, and numerical solution of stochastic differential equations.
Prerequisites: CS 1114, MATH 2114, CMDA 2006
CMDA 3606: Concepts and techniques from numerical linear algebra, including iterative methods for solving linear systems and least squares problems, and numerical approaches for solving eigenvalue problems. Ill-posed inverse problems such as parameter estimation, and numerical methods of computing solutions to inverse problems. Numerical optimization. Emphasis on large-scale problems.
Prerequisites: CMDA 3605
Survey of computer science concepts and tools that enable computational science and data analytics. Data Structure design and implementation. Analysis of data structure and algorithm performance. Introduction to high-performance computer architectures and parallel computation. Basic operating systems concepts that influence the performance of large-scale computational modeling and data analytics. Software development and software tools for computational modeling. Not for CS major credit.
Prerequisite: CS 2114 (C)
Note: This course is cross-listed with CS 3634
Basic principles and techniques in data analytics, including, what is meant by "learning from data;" methods for the collection of, storing, accessing, and manipulating standard-size and large datasets; data visualization; and identifying sources of bias. The concepts taught will be applied to real-world case studies.
Prerequisite: CMDA 2006 or equivalent
Note: This course is cross-listed with STAT 3654 and CS 3654
Introduction to partial differential equations, including modeling and classification of partial differential equations. Finite difference and finite elements methods for the numerical solution of partial differential equations including function approximation, interpolation, and quadrature. Numerical solution of nonlinear systems of equations. Uncertainty quantification, prediction.
Prerequisite: CMDA 3606
A technical analytics course that will teach supervised and unsupervised learning strategies, including regression, generalized linear models, regularization, dimension reduction methods, tree-based methods for classification, and clustering. Upper-level analytical methods are shown in practice: e.g., advanced naïve Bayes and neural networks.
Prerequisite: CMDA 3654
Note: This course is cross-listed with STAT 4654 and CS 4654
Stochastic modeling methods with an emphasis in computing are taught. Select concepts from the classical and Bayesian paradigms are explored to provide multiple perspectives for how to learn from complex, datasets. There is particular focus on nested, spatial, and time series models.
Prerequisite: CMDA 2006
Note: This course is cross-listed with STAT 4664
Capstone research project for Computational Modeling and Data Analytics majors. Cultivate skills including reviewing the literature, creative problem solving, teamwork, critical thinking, and oral, written, and visual communications.
Prerequisites: CMDA 3606, CMDA 3634, CMDA 3654