Snehalata Huzurbazar, Jessi Cisewski, Bailey Fosdick, and Xia Wang

Statistical and Applied Mathematical Sciences Institute (SAMSI) Deputy Director Snehalata Huzurbazar, former SAMSI graduate fellow Jessi Cisewski, SAMSI postdoctoral fellow Bailey Fosdick, and former National Institute of Statistical Sciences (NISS) postdoctoral fellow Xia Wang gave an overview of NISS and SAMSI, both located in Research Triangle Park (RTP), North Carolina. Drawing on their experiences, they informed the audience about the wide range of opportunities for involvement at the institutes. What follows is a synopsis of the content of their talks at the 2014 Women in Statistics Conference.

Snehalata Huzurbazar, 2012–2014 Deputy Director of SAMSI, on leave from the Department of Statistics, University of Wyoming

Huzurbazar began with a brief history of NISS and SAMSI, describing their inceptions and the connections between the institutes. Incorporated in 1990, NISS was created by the American Statistical Association, the International Biometric Society, the Institute of Mathematical Statistics, the three main Universities in the area—Duke, North Carolina State University (NCSU), and The University of North Carolina at Chapel Hill (UNC)—and the Research Triangle Institute; representatives of these institutions serve on the board governing NISS. Research at NISS is project driven, with funding coming from various governmental and industrial organizations. For junior researchers, the most common opportunities are as postdoctoral researchers hired on projects and mentored by senior-level statisticians at NISS. The location of most NISS researchers is either in RTP or in the Washington, DC, area.

SAMSI, founded in 2002, is the only one of the National Science Foundation’s (NSF) seven mathematical sciences institutes to explicitly house statistics. The NSF grant that funds SAMSI is awarded to the three partner universities; SAMSI’s fourth partner is NISS. Both SAMSI and NISS are physically located in the NISS building in RTP, and interactions between the two institutes range from shared staff to collaboration in research activities in the form of joint workshops. As an NSF research institute, SAMSI’s main activity is the development and implementation of up to two academic year–long research programs. Topics for these programs are chosen to be in line with SAMSI’s mission to forge a synthesis of the statistical and applied mathematical sciences with disciplinary sciences to tackle important scientific challenges.

SAMSI research programs are planned and developed by an organizing committee comprised of a SAMSI directorate member and leaders in the specific research area for the program; planning can take up to two years. Each program begins with a week-long opening workshop in either late August or early September, when up to 180–200 researchers participate in a program with overview tutorials, talks, and an elaborate poster session that leads to the formation of focused research groups, called working groups. Over the next nine months, research proceeds with the working groups meeting weekly at SAMSI; members at other locations participate via WebEx.

There are anywhere from 6–12 working groups per research program. SAMSI’s main role is to facilitate connections between researchers by providing infrastructure such as WebEx for meetings, office space for researchers, and funds for visitors to SAMSI and participants in SAMSI workshops. Over the course of the year, SAMSI offers up to two graduate courses as part of each research program; local graduate students can take these for credit, while visitors can simply attend. During the remainder of the program year, SAMSI also hosts shorter workshops lasting up to two and half days on focused topics relevant to the program, and the research programs end with a transition workshop, usually held in May. The opportunities for participation range from suggesting topics and submitting proposals for future research programs to serving on organizing committees and visiting SAMSI during the academic year of the research program.

For each program, SAMSI hires three postdoctoral associates with the deadline for application usually early in December; research topics of the programs do have to match up to some extent with the applicant’s interests and training. Graduate students also can participate in the programs; students can join and participate in WGs and, if their advisor is visiting SAMSI, funds are available for more senior PhD students to visit.

SAMSI also organizes, or sometimes co-organizes, summer workshops. For example, in summer 2014, there was a joint SAMSI-NCAR workshop on surface temperature data. These are shorter workshops, sometimes later leading to year-long research workshops. A recent example is the neuroimaging data analysis workshop from summer 2013, which led to a successful proposal for the 2015–2016 SAMSI research program on neurosciences.

Huzurbazar then discussed opportunities at SAMSI for undergraduate and graduate students. Each of the main research programs has an accompanying 1.5-day undergraduate workshop, for which SAMSI usually has enough funds to support travel and lodging for up to 30 students; these workshops take place in October and February. There is also a week-long undergraduate workshop in May on modeling topics. In 2013, students used a database from NCSU researchers to predict hurricanes; in 2014, students modeled social network data acquired by the MIT media lab. SAMSI also funds graduate students to attend its 10-day Industrial Mathematical and Statistical Modeling Workshop every July, where students work on projects from industry supervised by teams of industry and academic mentors.

As part of the NSF Math Institutes and their Diversity Committee, SAMSI is involved with organizing activities such as the Modern Math Workshop (SAMSI is the lead organizer for 2014), which takes place before the annual meeting of the Society for Chicanos and Native Americans in Science (SACNAS); the Spring Opportunities Workshop for Women in the Mathematical Sciences (SAMSI was a co-organizer with NIMBioS in 2014); and the Blackwell-Tapia Conference (SAMSI was the lead organizer in 2008). In addition, SAMSI hosts activities such as the Recruiting and Retaining Graduate Students in the Statistical Sciences and Applied Mathematics workshop (June 2014), co-sponsored by the Alliance for Doctoral Studies in the Mathematical Sciences, with the aim of enhancing success for U.S. doctoral students, especially those from under-represented minority groups. The undergraduate and graduate workshops, along with outreach activities to include members from under-represented groups—including women—were stressed, because they are essential for helping with the recruitment and training of future statisticians.

After giving the audience information about upcoming SAMSI programs—Beyond Bioinformatics and Mathematical & Statistical Ecology for 2014–2015 and Neurosciences and Forensics for 2015–2016—Huzurbazar spoke about her own experience at SAMSI, first as a visitor in the analysis of object data (AOD) program in Spring 2011 and later as deputy director. Specifically, she described participation in two working groups in the 2010–2011 AOD program and one in the 2012–2013 massive datasets (MD) program, describing the three different yet beneficial working group atmospheres. From the AOD working groups, she made connections that helped progress her own research, while the MD working group introduced her to an entirely new research area for which she has a manuscript in progress.

Huzurbazar ended by giving her own observations about what is potentially different for female researchers. First, she noted that, as a single parent, she found it impossible to think about visiting SAMSI until her daughter was old enough to be in school, as the prospect of moving and setting up child care for a longer visit seemed insurmountable. The AOD program, her research interests, and her daughter’s year of kindergarten all coincided in 2010–2011. In her application to be a visitor in AOD, she mentioned that due to parenting constraints, she could either visit for a semester or not at all; coming for a few weeks or a month was simply not feasible.

This is a situation facing many parents, and often proves to be a binding constraint for female researchers with children. She learned that SAMSI is open to working with such constraints, and that women should ask to arrange visits that will work for them if the advertised structure does not. She also stressed that since working group participation can function via WebEx, not being able to visit should not prevent involvement. Later, as deputy director and with the help of SAMSI staff, she helped put together a list of nearby child care centers for short-term visits. SAMSI now has facilities to be used as lactation rooms in both the NISS building and the NC Biotechnology Center, where larger workshops take place. Though NSF funds cannot be used to reimburse child care expenses, institutes might have other funds available.

Finally, SAMSI is always looking for new areas in which to organize programs. Writing a proposal for a SAMSI program is fairly simple, and women need to take the lead in sending in such proposals. Huzurbazar stressed that solid researchers, especially female researchers, should make themselves and their research known to SAMSI.

Jessi Cisewski, Visiting Assistant Professor, Department of Statistics, Carnegie Mellon University

Starting as a statistics PhD student in 2007 at The University of North Carolina at Chapel Hill, Cisewski didn’t realize she was going to be spending the next five years in a gem of a location. With SAMSI, Duke, and NC State nearby, there was always something exciting and statistical brewing. In 2008, during the SAMSI course on sequential Monte Carlo methods, Cisewski’s eyes were opened to the world of statistics beyond the borders of her department.

She continued to go back for many events, lectures, seminars, and workshops, each providing manifold opportunities to meet top researchers in diverse topics. During her final year at UNC, she was a SAMSI graduate student fellow for the program on uncertainty quantification. She typically spent one day each week at SAMSI to attend the postdoc seminars, working group meetings, and the evening lecture. She treasured that year for several reasons. First, she was able to see how researchers from different fields (in this case, applied math and statistics) could come together, mutually understand each other’s jargon, and then formulate and address problems that require both disciplines. Second, it was great to get a glimpse of what life is like in academics after finishing graduate school by spending time with the SAMSI postdocs. And finally, it is just all-around enjoyable to be at SAMSI. If any PhD student or new researcher has the opportunity to spend time at SAMSI for an entire program, workshop, or class, Cisewski highly recommends taking advantage of it.

Another way to take advantage of what SAMSI has to offer is through summer research programs. For three weeks in June of 2013, a number of statisticians and astronomers came together for the program on modern statistical and computational methods for analysis of Kepler data. This was a program designed to get statisticians and astronomers talking about statistical methods for analyzing a particular type of exoplanet data. (Exoplanets are planets orbiting stars outside our solar system.) This program was timely due to the release of NASA’s Kepler data in October of 2012. It was a success, and the participants have continued to meet weekly or biweekly. They held a follow-up workshop at Carnegie Mellon University in June 2014 dubbed “ExoStat2014.” This was one of a number of opportunities for statisticians to get involved in astrostatistics. Anyone interested in research related to astronomy, astrophysics, and cosmology can join the newly formed ASA Interest Group on Astrostatistics, a group that has at least some of its origins in programs at SAMSI.

Bailey Fosdick, Postdoctoral Fellow, SAMSI, and Assistant Professor, Department of Statistics, Colorado State University

In the fall of 2013, Fosdick started as a postdoctoral fellow at SAMSI affiliated with the year-long computational methods in the social sciences (CMSS) program. The work in her dissertation was motivated by network analysis problems in the social sciences, so this SAMSI program seemed like the perfect opportunity for her to continue growing expertise in this area. Upon completing her PhD in statistics from the University of Washington, she was a confident young researcher, ready to confront the greater world of statistics. Little did she know how much learning there was still left for her to do! Luckily, experiences at SAMSI were quick to point this out and provide her with an array of opportunities to learn and grow.

Looking back over this past year, Fosdick says her experiences at SAMSI have been incredibly beneficial: She saw tremendous growth in her professional network, gained exposure and experience in a number of new research areas, and broadened her perspective and understanding of science appreciably.

Throughout the past year, Fosdick attended seven SAMSI workshops, where she met numerous statisticians working on research problems similar to her own and problems she had never considered, or even heard of, before. Already, a few of these connections have formed into mentoring relationships and research collaborations.

Even though Fosdick was a postdoctoral fellow associated with the CMSS program, she was encouraged to participate in the low-dimensional structure in high-dimensional systems (LDHD) program that was occurring simultaneously. Arriving at SAMSI, she knew she wanted to take advantage of this invitation. She attended both graduate-level courses in the fall, which provided gentle introductions to the foundational work and methods used in the program research areas. She also became involved in various working groups in both programs and has since started work in two new research areas that she had no experience in before arriving at SAMSI.

SAMSI brings applied mathematicians and statisticians together to confront problems in science. Interacting with applied mathematicians in working groups and on projects introduced Fosdick to new perspectives on approaching research problems. The assimilation of research strategies she gathered will benefit her in years to come as she continues to push the boundaries of science.

As a young researcher, Fosdick was able to explore new areas and gain a greater understanding of, appreciation for, and familiarity with not only the domain of statistics, but also the role of the mathematical sciences across many fields of science.

Xia Wang, Assistant Professor, Mathematical Sciences, University of Cincinnati

Wang earned her PhD in statistics from the University of Connecticut in 2009. Right after graduation, she joined NISS as a postdoctoral fellow. The two-year training at NISS provided a big boost to her career in academia.

NISS’s postdocs work on specific projects under mentoring of senior directors. Wang was on the NCI-funded project Proteomic Technologies: Metrology, Research Methods Design, and Analysis. The research group included two NISS senior directors and three postdocs with specific statistical or bioinformatics expertise. Wang’s job as a postdoc was to provide statistical support for multiple institutions participating in the NCI clinical proteomic technology assessment for cancer program. With the mentorship of Nell Sedransk, she started working with researchers in biotechnology and proteomics. It was stressful at the beginning, said Wang, but ultimately rewarding. It was during the years at NISS that she observed and learned how to effectively communicate with people from different disciplines.

The collaborations Wang established at NISS continued after she joined the University of Cincinnati as an assistant professor in statistics. Being a NISS postdoc, she was encouraged to participate in local research activities, including workshops, seminars, and conferences. There are many! She benefitted the most from the yearly SAMSI programs. With SAMSI in the same building as NISS, Wang was able to sit in on any working group’s meetings, workshops, or courses whenever she was able. She was exposed to a wide collection of interesting topics during those two years.

SAMSI also provided strong staff support, a welcoming physical environment, and technologies that help people communicate—both formally and informally. While a NISS postdoc, Wang participated in a working group in the 2009–2010 SAMSI space-time analysis program. This working group has been working together for the past five years. The familiarity with the SAMSI program procedure encouraged her to remotely participate in a SAMSI 2012–2013 program through WebEx. In the middle of the year, three junior faculty, all from different institutes, started working together on a project. All had different research interests. Wang calls their collaboration a “SAMSI product,” as it is hard to imagine how the three could or would have started collaborating without the venue provided by SAMSI.

Wang cherishes the unique opportunities NISS offers its postdocs and recommends it to all new graduates or junior researchers considering postdoc opportunities.

More information can be found at the NISS and SAMSI websites.

About the Authors

Jessi Cisewski is a visiting assistant professor in the department of statistics at Carnegie Mellon University. She earned her PhD in statistics in May of 2012 from the department of statistics and operations research at The University of North Carolina at Chapel Hill. Cisewski was a Statistical and Applied Mathematical Sciences Institute graduate fellow during the 2011–2012 program on uncertainty quantification and has participated in several SAMSI programs.

Bailey Fosdick completed her PhD last year in the department of statistics at the University of Washington. She is a postdoctoral fellow at the Statistical and Applied Mathematical Sciences Institute and Duke University, but will soon join the faculty in the department of statistics at Colorado State University. Her primary research interests lie in covariance models for multi-way data, social network analysis, and applications of Bayesian methodology in the social sciences.

Snehalata Huzurbazar is an associate professor at the University of Wyoming and served as deputy director of Statistical and Applied Mathematical Sciences Institute from 2012–2014. She holds a PhD in statistics from Colorado State University. Her methodological research interests lie in saddlepoint approximations, grain-size distributions, functional data analysis, and survival analysis. Her application areas include evolutionary bioinformatics, sedimentology, soil science, glaciology, and wildlife biology.

Xia Wang is an assistant professor in the department of mathematical sciences at the University of Cincinnati. She earned her doctoral degree in statistics from the University of Connecticut in 2009 and worked as a postdoctoral fellow at the National Institute of Statistical Sciences from 2009–2011. Her research interests include Bayesian methodology and computation, categorical data analysis, spatial statistics and spatial-temporal statistics, and applications of statistical models in genomics and proteomics data.

Back to Top