The Interdisciplinary Discussion Course (IDC)


The IDC meets Mondays 6-7:30pm at Market Place Upper East. These sessions will generally feature a guest speaker and dinner will be provided.


Week 1 (Aug 25):

Introduction to ‘What If?’

Speakers: Mine Cetinkaya-Rundel, Anita Layton, Xiaobai Sun, Nick Gessler


Week 2 (Sep 1):

Is prison contagious?

Kristian Lum

Kristian Lum
DataPad

Abstract: This talk will discuss an agent-based model of incarceration based on the susceptible–infected–susceptible (SIS) model of infectious disease propagation. Our central hypothesis is that the observed racial disparities in incarceration rates between Black and White Americans can be explained as the result of differential sentencing between the two demographic groups. We demonstrate that if incarceration can be spread through a social influence network, then even relatively small differences in sentencing can result in large disparities in incarceration rates. Controlling for effects of transmissibility, susceptibility and influence network structure, our model reproduces the observed large disparities in incarceration rates given the differences in sentence lengths for White and Black drug offenders in the USA without extensive parameter tuning. We further establish the suitability of the SIS model as applied to incarceration by demonstrating that the observed structural patterns of recidivism are an emergent property of the model. In fact, our model shows a remarkably close correspondence with California incarceration data. This work advances efforts to combine the theories and methods of epidemiology and criminology.

About the speaker: Kristian Lum is a Data Scientist at DataPad. Dr. Lum previously worked with the Center for Peace Studies and Violence Prevention at Virginia Tech. She received her PhD in Statistical Science from Duke University.

Prepare for the talk: Read Is prison contagious? (Science Magazine).


Week 3 (Sep 8):

Modeling the equilibria

Scott de Marchi

Scott de Marchi
Duke University, Political Science

Abstract: In this talk Dr. de Marchi will discuss equilibria in free markets and how to model them.

About the speaker: Scott de Marchi is Professor and Associate Chair of Political Science at Duke University. Dr. de Marchi’s work focuses on mathematical methods, especially computational social science, machine learning, and mixed methods. Substantively, he examines individual decision-making in contexts that include the American Congress and presidency, bargaining in legislatures, interstate conflict, and voting behavior. He has been an external fellow at the Santa Fe Institute and the National Defense University and is currently a principal investigator for NSF’s EITM program.


Week 4 (Sep 15):

A look back at the history of statistics and genetics

Sayan Mukherjee

Sayan Mukherjee
Duke University, Statistical Science

Abstract: In this talk Dr. Mukherjee will discuss the parallel history of the development of statistics and genetics from 1885 to the 1940’s and the influence they had on each other.

About the speaker: Sayan Mukherjee is an Associate Professor of Statistical Science, Computer Science, and Mathematics at Duke University. Dr. Mukherjee’s research interests include geometry and topology in probabilistic modeling. Statistical and computational biology. Modeling of massive data.


Week 5 (Sep 22):

Classifying Musical Genres:

An investigation into Sorting Music using Topology

Paul Bendich

Derrick Nowack

Marshall Ratliff

Marshall Ratliff, Derrick Nowack, Paul Bendich
Duke University, Mathematics

Abstract: This project involved some typical data analytic methods such as dimensionality reduction with principal component analysis and a variety of machine learning techniques, however the core of this work is based on applying very new methods in computational topology coming out of Duke, in particular 1-dimensional persistent homology. We will give a very brief but comprehensive crash course in persistent homology for background.

About the speakers: Marshall Ratliff is a senior (Class of 2015) working toward a B.S. in mathematics and minor in finance. He has done work in Cold Storage and Logistics consulting and his primary interests are in the (rather different) fields of number theory and data sciences.

Derrick Nowack is a sophomore (Class of 2017) planning on majoring in Mathematics with a minor in Polish.

Paul Bendich is an Assistant Research Professor in the Mathematics Department at Duke Unversity, as well as the Associate Director for Undergraduate Research of the Information Initiative at Duke (iiD). The aim of his research is to develop algebraic and topological tools for application in a wide variety of scientific areas, particularly the analysis of complex and high-dimensional datasets, and to find useful connections between computational topology and statistical methodology. At iiD, he directs several programs that aim to foster data-driven, interdisciplinary undergraduate research across the university.


Week 6 (Sep 29):

What to expect and what’s next: ‘What If?’

Kay Palopoli

Nettie Song

Speakers: Kay Palopoli & Nettie Song

Abstract: In this talk Kay and Nettie, former students of the What If? FOCUS cluster, will discuss their experiences in the cluster, as well as projects they have been working on since.

About the speakers: Yitian Song (Nettie) is currently a sophomore at Duke. Nettie will major in mathematics and considering having another major in computer science or minor in French. Kay is currently a sophomore at Duke, majoring in Biomedical Engineering and pursuing a Genome Sciences and Policy Certificate. This past summer, she was a fellow at Genome Institute and is continuing her work in the lab through independent study.


Week 7 (Oct 6):

Moving Fluid in Tubes: How do we use mathematics to help understand circulatoru systems in small organisms?

Austin Baird

Austin Baird
Duke University, Mathematics

Abstract: The human heart is a multi chambered organ using valves and compression to create pressure heads, which drive blood flow. Some organisms don’t share this type of heart structure, namely, very small invertebrates and embryonic vertebrates. We begin by discussing scaling in fluid dynamics and why merely existing at such a small scales can cause difficulty in effectively transporting fluid. We then move to describing effective pumping mechanisms in valveless tubular hearts and how numerical modeling can help understand how these organisms are able to transport blood, despite their limitations. Once we have a numerical framework to investigate these organisms we can then begin to add more biological structure in our models.

About the speaker: Austin Baird is a Visiting Assistant Professor in the Department of Mathematics at Duke University.


Week 8 (Oct 13): Fall Break

No meeting.


Week 9 (Oct 20): [TBA]

iiD

Robert Calderbank

Robert Calderbank
Duke University, iiD

Abstract: In this talk Dr. Calderbank will discuss the foundation of the Information Initiative at Duke (iiD) as well as highlight iiD programs and resources that may be of particular interest to undergraduate students.

About the speaker: Robert Calderbank is Professor of Computer Science, Electrical Engineering, and Mathematics at Duke University, as well as the Director of the Information Initiative at Duke.


Week 10 (Oct 27):

The Functional Anderson-Darling Test

Gina-Maria Pomann

Gina-Maria Pomann
NCSU, Department of Statistics

Abstract: A number of magnetic resonance (MR) imaging modalities can be used to measure the diffusion of water in the brain. An important question is which of these modalities are most useful for differentiating between MR images of patients with multiple sclerosis (MS) and those of healthy controls. We propose a hypothesis test that reduces the dimension of the testing problem in a way that enables the application of traditional nonparametric univariate tests. This results in a procedure that is computationally inexpensive. Simulation studies are presented to demonstrate the strength and validity of our approach. We also provide a comparison to a competing method. The proposed test is then illustrated by applying it to a state-of-the art diffusion tensor imaging(DTI) study where the objective is to compare white matter tract profiles in healthy individuals and multiple sclerosis (MS) patients.

About the speaker: Gina-Maria Pomann is a PhD student in the North Carolina State University Statistics Department. She is a National Science Foundation and AT&T Research Labs Fellow. Her current research focuses on developing novel statistical methodology to study disease progression of multiple sclerosis using magnetic resonance imaging techniques.


Week 11 (Nov 3):

[No IDC this week]


Week 12 (Nov 10):

Sex, Lies, and BMJ

Amy Herring

Amy Herring
UNC, Biostatistics

Abstract: The National Longitudinal Study of Adolescent Health (Add Health) provides a wealth of confidentially-reported data on the developmental transitions from adolescence to adulthood. We report results from two innovative projects based on data from Add Health. We start with a longitudinal Bayesian copula factor model for complex categorical data on sexual identity, sexual attraction, and partnering behavior of sexual minority youth and young adults, addressing novel questions of how these components of sexuality evolve in a critical developmental time window. We end with a holiday diversion examining modern self-report of virgin birth. These projects are in collaboration with Carolyn Halpern (UNC), Tsuyoshi Kunihama (Duke), David Dunson (Duke), Penny Gordon-Larsen (UNC), Samantha Attard (UNC), and Bill Joyner (The Chapel of the Cross).

About the speaker: Amy Herring is Associate Chair and Professor of Biostatistics at UNC. Dr. Herring’s research focuses on developing semiparametric Bayesian hierarchical models for highly correlated exposures, exposures to mixtures, and multivariate outcomes; developing statistical methods for missing and mismeasured exposure data; applications to environmental and reproductive epidemiology. Research funding includes NIEHS 1R01ES020619 (PI) and numerous collaborative projects dealing with birth defects, environmental and occupational exposures, obstetrics and gynecology, child neurodevelopment, adolescent development, and maternal health, including the National Children’s Study and National Birth Defects Prevention Study. Training funding includes NIEHS T32ES007018 (PI), a multidisciplinary training program in environmental biostatistics, environmental epidemiology, and environmental health sciences.

Prepare for the talk: Read the LATimes coverage of the associated article.


Week 13 (Nov 17):

Statistics and Machine Learning in RTP startups

Andrew Cron

Andrew Cron
Weinraub Analytics

Abstract: Andrew will discuss his experiences as a Data Scientist in RTP startups in various fields including Web Tech and Finance.

About the speaker: Andrew Cron is the CTO of Weinraub Analytics, a start-up in Durham, NC which aims to provide quantitative money management solutions to maximize the return on capital through rigorous scientific research and decision analysis.


Week 14 (Nov 24):

Analyzing first flowering event data using survival models with space and time-varying covariates

Maria Terres

Maria Terres
NCSU, Statistics

Abstract: First flowering events in cherry trees are believed to be closely related to temperature patterns during the winter and spring months. Earlier works have incorporated the idea of temperature thresholds, defining chill and heat functions based on these thresholds. However, selection of the thresholds is often arbitrary and shared across species and locations. We propose a survival model with spatially and temporally varying covariates having functional forms representing chill and heat accumulation leading up to first flowering events. Thresholds are chosen utlizing the ranked probability scores, selecting the threshold pair that minimizes the difference between the predicted and observed cumulative probability curves. We first apply the model using temporally varying covariates to analyze 29 years of flowering data for four cherry species (Cerasus spp.) grown in Hachioji, Japan. This allows us to investigate whether relationship with temperature may vary between earlier and later flowering species. Next, the model is applied to 52 years of flowering data for 45 Cerasus spachiana × C. speciosa trees grown across Japan’s Honshu Island using spatially and temporally varying covariates and spatial random effects. By exploring flowering dates across locations, we can explore how the relationship between temperature and first flowering events varies through space.

About the speaker: Maria A. Terres is a Postdoctoral Research Scholar at North Carolina State University working with Prof. Montserrat Fuentes. She received her Ph.D. in Statistical Science from Duke University working with Prof. Alan E. Gelfand. She holds a B.A. in Biology and Mathematics from Bard College at Simon’s Rock and an M.A. in Statistics from Columbia University. Her research focuses on Bayesian spatial-temporal modeling for environmental and ecological applications.


Week 15 (Dec 1):

Mosquitoes and subjective expectations:

Designing a human experiment in Kenya

Liz Turner

Liz Turner
Duke University, Global Health Institute

Abstract: In this talk Liz Turner will discuss an ongoing NIH-funded study in Western Kenya that seeks to improve targeting of anti-malarial drugs to those who need them.She will introduce the problem, talk about the importance of economics (and of economists, epidemiologists and statisticians working together) and how to move from a research question to generating important pilot data from field work to implications for designing and implementing a large-scale human experiment. The work is ongoing and still in the piloting phase so that provides a good opportunity to design current issues in the life of a study.

About the speaker: Liz Turner is Assistant Professor of Biostatistics and Bioinformatics and Global Health at Duke University. Dr. Turner’s work focuses on the design, implementation and analysis of cluster-randomized trials, with a particular focus on studies in low and middle-income countries. She is faculty in Duke’s Biostatistics Department and Duke Global Health Institute (DGHI). She leads the Duke Global Health (DGHI) Biostatistics Core, whose mission it is to collaborate with DGHI faculty from departments across campus, and has worked on projects as far a field as East Africa and China.