Bucknell Mathematics & Statistics Department Blog

Title: Piecewise Deterministic Markov Processes and Metal Grain Coarsening

Abstract: In material science, individual metal grains obey surprisingly simple rules.  However, geometric considerations can create difficulties when attempting to analyze the bulk properties of metals.  In this talk, we’ll describe a mean field model using piecewise deterministic Markov processes which converts the geometric problem of grain evolution to one of analysis and stochastic processes.  In particular, we’ll show that densities of grains approach a law of large numbers described by a partial differential equation.

Title: How Analytics Professionals Make the World Go Round

Abstract:

It’s hard to open a business magazine, walk through an airport or even participate in an executive meeting and NOT see or hear reference to analytics. Recognized as the currency of business, mathematics or analytics are empowering decision-making at new levels. A trend that started with CIOs is spreading throughout the C-Suite – Everyone wants ANALYTICS!

What exactly are these analytics professionals doing? What types of real-world problems are they solving? Are they actually applying the math, statistics and operations research tools they learned in their courses? What other skills or certifications do they have?  What opportunities are there for graduate studies in analytics?
Hear from Anne Robinson and John Williamson of Verizon Wireless.  Anne Robinson holds a Ph.D. in Industrial Engineering from Stanford University and served as President of INFORMS, the largest professional organization for both practitioners and academics in the field of Analytics/Operations Research.  John Williamson holds a M.B.A. in Supply Chain Management from Penn State University.  There will be an opportunity to speak with the presenters following the talk.

Title: Hierarchical Social Network Models with Applications in the Social Sciences

Abstract: The term “social network analysis” includes most quantitative methods for analyzing relational data and these methods are both exploratory and inferential. For inference, there are a variety of social network statistical models to represent the stochastic nature of network relationships.  In this talk, I will present social network models that are written as Hierarchical Bayes models and describe how these generative models can be used in the social sciences which often involve multiple partially exchangeable networks.