Research

Research Interests:
Bayesian Statistics - specifically change point models and variable selection techniques, as well as applications to computational biology.

The primary focus of my research has been the identification change points in climatic time series.  A change point acts as a regime boundary, where the climate system changes from one state to another.  As an example, my Ph.D. research looked at a 5 million year proxy record (Lisiecki & Raymo 2005 ) of global ice volume. This data set gives evidence for at least two changes in glacial dynamics.  Around 2.7 million years ago more permanent glaciers began to form in the Northern Hemisphere, evidenced by an increase in the amplitude of the proxy record.  More recently, around 1 million years ago, not only was there a further increase in the amplitude of the proxy record, but the frequency of glacial melting events changed from every 40,000 years to ~100,000 years.  A more recent example can be found in the global surface temperature anomalies data set produced by NOAA (NOAA website), which is freqeuntly cited in the context of global warming

The goal of change point analysis is to provide uncertainty estimates both in the number and timing of change points in these climatic records, but to do so in a computationally efficient way.  Given a data set with n observations and a desire for k change points, there are nCk potential solutions to the multiple change point problem. Thus, a brute force attempt to study all possible placements of a specified number of change points quickly becomes infeasible, as the number of possible solutions grows exponentially in the length of the data set.  To date, least squares and Bayesian change point methods have been developed for regression models, as well as an algorithm that allows for variable selection to help distinguish between competing hypotheses. Recent work has focused on a sequential approach to chagne point detection and the development of an algorithm that can quickly update its inference with each additional observation. In addition, an even more efficient, but approximate algorithm has been developed whose solution compares favorably to the exact Bayesian algorithms developed to date. Next, our plan is to generalize the model so that its hyperparameters can be learned from the data (rather than being specified by the user) and also develop a model that can incorporate a correlated error structure.

As a hobby, I love sports statistics because it combines two of my favorite things, sports and statistics! I've really enjoyed working on these types of projects over the last several years with some of our undergraduate students.

 

Publications:

• Ruggieri, E., Yu, M., and Qiang, R. (2020), “Change Point Models: What are They and Why Do We Need Them?”, Undergraduate Mathematics and Its Applications 41(1), 61-80.
• Yu, M. and Ruggieri, E. (2019), “Change Point Analysis of Global Temperature Records, ” International Journal of Climatology 39(8), 3679-3688. doi: 10.1002/joc.6042
• Ruggieri, E. (2018), “A Pruned, Recursive Solution to the Multiple Change Point Problem,” Computational Statistics, 33(2), 1017-1045. doi: 10.1007/s00180-017-0756-9
• Civitarese, AM, Ruggieri, E., Walz, JM, Mack, DA, Heard, SO, Mitchell, M., Lilly, C.M., Landry, K.E., and Ellison, R.T. (2017), “A 10-Year Review of Total Hospital-Onset ICU Bloodstream Infections at an Academic Medical Center,“ CHEST, 151(5), 1011-1017. doi: 10.1016/j.chest.2017.02.008
• Ruggieri, E. and Antonellis, M. (2016), “An exact approach to sequential Bayesian change point detection,” Computational Statistics and Data Analysis 97, 71-86. doi: 10.1016/j.csda.2015.11.010
• Caramiciu, J.A., Adams, J.P., McKnown, B.T., Frence, C.D., Ruggieri, E.R. and Heard S.O. (2014), “Effects of an in-house coordinator and practitioner referral rather than proxy referral on tissue donation rates,” Transplantation Proceedings, 46(5), 1274-1280. doi: 10.1016/j.transproceed.2014.03.005
• Ruggieri E. and Lawrence, C.E. (2014), “The Bayesian Change Point and Variable Selection Algorithm: Application to the δ18O Record of the Plio-Pleistocene,” Journal of Computational and Graphical Statistics , 23 (1), 87-110. doi:10.1080/10618600.2012.707852
• Ruggieri&Lawrence 2014 (pdf)
• Ruggieri, E. (2013), “A Bayesian Approach to Detecting Change Points in Climatic Records,” International Journal of Climatology. 33(2) 520-528. doi: 10.1002/joc.3447
  • Ruggieri2013 (pdf)
• Ruggieri, E. and Lawrence, C.E. (2012), “On Efficient Calculations for Bayesian Variable Selection,” Computational Statistics and Data Analysis, 56, 1319-1332. doi:10.1016/j.csda.2011.09.026
  • Ruggieri&Lawrence2012 (pdf)
• Ruggieri, E. (2011), “Inference in discrete high dimensional space: An exploration of the Earth’s ice sheets through change point and variable selection techniques,” Ph.D. Dissertation, Brown University.
• Ruggieri, E., Herbert, T., Lawrence, K., and Lawrence, C.E. (2009), “The Change Point Method for Detecting Regime Shifts in Paleoclimatic Time Series: Application to δ18O Time Series of the Plio-Pleistocene,” Paleoceanography, 24, PA1204, doi:10.1029/2007PA001568
  • Ruggieri_et_al_2009 (pdf)
• Ruggieri, E. and Schreiber, S.J. (2005), “The Dynamics of the Schoener-Holt Model of Intra-Guild Predation,” Mathematical Biosciences & Engineering, 2, 279-288. PubMed link
  • Ruggieri&Schreiber2005 (pdf)

 

Conference Proceedings

• Civitarese, A., Ellison, R.T., Mack, D.A., Ruggieri, E., Heard, S.O., and Walz, J.M. (2014), “Sustained Reduction in Nosocomial Bloodstream Infections in the ICU Setting,” Critical Care Medicine, 42(12), A1477. doi: 10.1097/01.ccm.0000457983.02079.c0

 

Letter to the Editor

• Civitarese, AM, Ruggieri, E., Walz, JM, Mack, DA, Heard, SO, Mitchell, M., Lilly, C.M., Landry, K.E., and Ellison, R.T. (2019), “RE: Preventability of hospital onset bacteremia and fungemia: A pilot study of potential healthcare-associated infection outcome measure, by Dantes et al (2019)”, Infection Control and Hospital Epidemiology 40(10), 1209-1210. doi: 10.1017/ice.2019.193

 

Works in Progress

• with I.G. Hatvani, D. Topal, and Z. Kern. (in progress) “Spatially clustered changepoints in ice core d18O across Greenland concurring with regime-shifts of large-scale atmospheric variability over the past millennium”.
• The Mid-Pleistocene Transition: From Forcing to Pacing of Ice Sheets (in progress)

 

Presentations

• Change Point Analysis of Global Temperature Records, Contributed Paper Session, MAA Northeast Section Spring Meeting, June 2019
• Climate Change, GPS Directions, and DNA Searches: How Recursion Can Make Hard Problems Easier to Solve, Pi Mu Epsilon Induction Ceremony, Providence College, April 2017
• Climate Change! Global Warming! Searching for Change Points in Climate Records, Assumption College, Natural Science Seminar Series, February 2016
• A Pruned Recursive Solution to the Multiple Change Point Problem, AMS Session on Probability Theory, Stochastic Processes and Statistics, Joint Mathematics Meetings, Seattle, WA, January 2016
• A Bayesian Approach to Sequential Change Point Detection, AMS Session on Statistics, Joint Mathematics Meetings, San Antonio, TX, January 2015
• A Sequential Approach to Detecting Change Points, AMS Session on Statistical Modeling, Big Data, and Computing, Joint Mathematics Meetings, Baltimore, MD, January 2014
• A Bayesian Approach to Detecting Change Points in Climatic Records, New Faculty Talk, MAA Northest Fall Section Meeting, Wheaton College, Norton, MA, November 2013
• A Bayesian Approach to Detecting Change Points in Climatic Records, Department of Mathematics and Computer Science Colloquium, Providence College, Providence, RI, October 2013
• A Bayesian Approach to Detecting Change Points in Climatic Records, College of the Holy Cross, Worcester, MA, January 2013
• A Bayesian Approach to Detecting Change Points in Climatic Records, AMS Session on Probability and Statistics, Joint Mathematics Meetings, San Diego, CA, January 2013
• A More Efficient Approach to Bayesian Variable Selection, AMS Special Session on Groups, Algorithms, Complexity, and Theory of Security, Joint Mathematics Meetings, Boston, MA, January 2012
• A Mathematical Exploration of the Earth’s Glacial System, Pi Mu Epsilon Induction Ceremony, Providence College, Providence, RI, April 2010.
• The Frequency of Glacial Events, University of Massachusetts at Amherst, Amherst, MA, February 2010
• The Frequency of Glacial Events, Kenyon College, Gambier, OH, February 2010
• The Frequency of Glacial Events, Niagara University, Niagara, NY, February 2010
• The Frequency of Glacial Events, Duquesne University, Pittsburgh, PA, February 2010
• The Mid-Pleistocene Transition: From Forcing to Pacing of Ice Sheets, AMS Session on Probability and Statistics, Joint Mathematics Meetings, San Francisco, CA, January 2010