Title: A McNemar’s-like Odds Ratio and Test for Multivariate Paired Binary Data.
Student Talk Series: Ravi Ramakrishna, April 7th @ 12 noon in Olin 266
Title: “Die ganzen Zahlen hat der liebe Gott gemacht, alles andere ist Menschenwerk.” – L. Kronecker. Translation: “God made the integers. All else is the work of Man.”
Abstract: We were taught about the number line and its properties in middle school. Only in college do we learn that the number line is created by taking the rationals and “filling in the holes”. (It took mathematicians 200+ years to understand all this, so the 8 or so years it’s taking you is fast.) In this talk we’ll explore other ways of “filling in the Holes” and why these other number systems are both natural and powerful things.
Student Talk Series: Matt Mizuhara ’12, March 24th @ 12 noon in Olin 268
Title: Mathematical biology under the microscope: A study of cell motility
Abstract: Although physics and chemistry have long relied on mathematics as a descriptive and exploratory tool, biological systems were historically seen as too complex to be understood theoretically. However, advances in mathematics and computational capabilities now allow for the quantification of biological problems in a field called mathematical biology.
In this talk I will introduce a modern topic of mathematical biology: crawling cell motility. Cell motion plays a central role in wound healing and the immune response, e.g., to fight foreign bodies. We will present a partial differential equation model for cell motion proposed by Ziebert et al. (2011). The subsequent analytical and numerical studies give rise to surprising mathematical results as well as novel insights for biologists, including applications to directed cell motility and sorting. This talk will not require any prerequisite knowledge of partial differential equations or biology, though a calculus background will be helpful.
Student Talk Series: Lara Dick, February 25th @ 12 noon in Olin 268
Title: There’s Nothing Common About It
Abstract: Training yourself to think like a child is often counterintuitive and surprisingly difficult. My research focuses on helping preservice teachers analyze and learn from children’s mathematical thinking. For this talk, we will focus on different types of story problems as described in the Common Core State Standards for Mathematics and discuss how children change their solution approaches depending on the situation presented in the problem. We will watch videos of children solving different story problems and will consider various samples of written student work. Audience participation will be encouraged.
Student Talk Series: Greg Adams, February 11th @ 12 noon in Olin 268
Title: Monkeying around in South Africa
Abstract: As part of my sabbatical last year, I spent 5 weeks in South Africa at the Unizulu Science Centre in Richards Bay. One of the highlights of my stay were the many encounters with vervet monkeys, both in the flesh and abstractly in mathematical problems. They insist on playing a role in this talk.
The science center serves the some 700 rural schools in the province of Kwazulu-Natal. The level of mathematics achievement by the students at the rural public schools in South Africa is among the lowest in the world. The science center was charged by the Department of Education to conduct teacher training in geometry for the school teachers in the province.
During my talk, I will discuss my past and future role in the initiatives of the science center and will talk about some of my experiences
Student Talk Series: Mark Meyer, January 28th @ 12 noon in Olin 268
Title: I’m all about that Bayes, ’bout that Bayes.
Abstract: A September 2014 New York Times article titled “The Odds, Continually Updated” discusses the growing popularity of Bayesian statistics both within and outside of the statistical community. This expansion is due in part to the growth of computing power over the last decade and a half. So what is Bayesian statistics? The title of the article, and indeed the article itself, suggest that Bayesian statistics uses, even requires, prior information to inform the analysis. But this is only a small aspect of the Bayesian approach. We can use Bayesian statistics to analyze any data and, as we shall see, it can even provide more informative solutions than the Frequentist, or classical, approach to many problems. This talk will discuss the two philosophies of statistics: Bayesian and Frequentist. In doing so, we will cover some history behind the Bayesian paradigm, introduce the general approach to Bayesian statistics, and discuss several real examples, in each case comparing the Bayesian and Frequentist approaches to each other.
DVP Talk: Sam Ventura, 12/7 @ 4 pm in Olin 264
Title: Classification and Clustering for Record Linkage in Large Datasets
Abstract: Record linkage, or the process of linking records corresponding to unique entities within and/or across data sources, is an increasingly important problem in today’s data-rich world. Due to issues like typographical errors, name variation, and repetition of common names, linking records of unique entities within and across large data sources can be a difficult task, in terms of both accuracy and computational feasibility. We frame record linkage as a clustering problem, where the objects to be clustered are the records in the data source(s), and the clusters are the unique entities to which the records correspond. We use the following three-step approach for record linkage: First, records are partitioned into blocks of loosely similar records to reduce the comparison space and ensure computational feasibility. We propose a sequential blocking approach that iterates through a nested set of decreasingly strict blocking criteria to reduce the comparison space more efficiently. Second, we adopt an ensemble supervised learning approach to estimate the probability that a pair of records matches. We propose a new adaptive prediction approach for classifier ensembles (specifically, random forests) that extracts and incorporates summary statistic information from the distribution of estimated probabilities. Third, after transforming our estimated pairwise probabilities of matching to pairwise dissimilarities, we use hierarchical clustering to link matching records. We apply these approaches to two labeled record linkage datasets: a set of labeled inventors from the United States Patent and Trademark Office database and a compilation of lists of death records from the Syrian Civil War conflict.
Student Talk Series: Sam Ventura, December 3 @ 12 noon in Olin 268
Title: A Crash-Course in Sports Statistics
Abstract: The path to becoming successful in the sporting world is substantially different from that of more traditional fields. Whether you’re an athlete or an analyst, coursework and university degrees matter much less being able to effectively demonstrate you ability to have a positive impact on a team or organization. In this talk, I will discuss my path to becoming successful in the world of statistics in sports. I will give a crash-course in the fundamentals of using statistical analysis in sports, discussing important concepts such as repeatability, replacement, and prediction. I will also touch on more academic topics that are extremely important in the sporting world, such as programming techniques, data visualization, professional communication (writing and speaking). Finally, I’ll discuss my personal experiences from working with professional sports teams, coaching ice hockey, and teaching students about statistics. Although my work is primarily in hockey, I will also discuss the use of statistics and quantitative analysis in football, baseball, and basketball.
MAA Student Event: Thursday 11/12 @ 6 in ACWS 108
Join the MAA for cookie decorating and a viewing of Moneyball, with Brad Pitt and Jonah Hill.
Cookie decorating (and eating!) begins at 6:00 followed by a talk on softball statistics by Coach Courtnay Foster, coach of the Bucknell’s Softball Team.
Coach Foster will talk at 6:30pm, and the movie begins at 6:45pm.
Sponsored by the Mathematics Department and the MAA Club
Student Talk Series: Shira Mitchell, October 29 @ 12 noon in Olin 268
Title: Causal inference for the Millennium Villages Project
Speaker: Dr. Shira Mitchell, Columbia University, Columbia Population Research Center
Abstract: The Millennium Villages Project (MVP) is a ten-year integrated rural development project implemented in ten sub-Saharan African sites. We describe the design for causal inference about the MVP’s effect on a variety of development indicators. Causal inference for the MVP context presents many challenges: a nonrandomized design, limited baseline data for candidate control areas, and the assignment of treatment to only ten sites, limiting effective sample sizes. We develop and carry out a matching procedure tailored to small samples and designed to facilitate communication with subject-matter experts. Following the design, we propose hierarchical Bayesian causal models for multiple outcomes. This work provides a case study of the careful evaluation design of a non-randomized intervention, with clear pre-specification of the procedure and matches before outcome data are available.