Arman Sabbaghi

Arman Sabbaghi

Assistant Professor, Applied Statistics (Harvard University; Statistics; 2014)

Dr. Sabbaghi has a long history in statistics, beginning with a Bachelor of Science degree from Purdue University in both pure mathematics and statistics, and continuing to a Ph. D.from Harvard University.  Initially, he was interested in time-series analysis, but at Harvard he was introduced to applications of experimental design in engineering, medicine, and social science.

Design of experiments focuses on planning a study to meet the specified objectives of the experimenter in order to ensure that the right type of data and sufficient sample size and power are available to answer the research questions of interest.  In applying experimental design techniques to engineering problems and seeing the exact outcomes in real-life applications, Arman found the focus of his research.

While at Harvard, Arman collaborated with his co-advisor Professor Tirthankar Dasgupta, and with Professor Qiang Huang, and his graduate students from the University of Southern California’s Daniel J. Epstein Department of Industrial and Systems Engineering, to better understand quality control for 3D printing.  Drawing on concepts from causal inference, he helped design experiments and performed Bayesian analyses to demonstrate a previously unrecognized interference issue, which has important implications for 3D printing in practice, in standard quality control techniques.

For Dr. Sabbaghi, returning ‘home’ to Purdue was a natural progression for his career.  With Purdue’s strong engineering program and Arman’s interest in applying his expertise in experimental design, causal inference, and Bayesian statistics it was a perfect research fit.  An added benefit to joining Purdue Statistics is that Arman is able to be closer to family.  Dr.Sabbaghi’s interest in being a member of a strong, broad, modern department of statistics with interdisciplinary collaborations catalyzed his return to Purdue Statistics.  Along with his work with 3D printing, Arman also focuses on statistical process/quality control in advanced manufacturing, the design and analysis of experiments and observational studies accounting for real-life complications, and powerful multiple comparison procedures of treatment effects that satisfy regulatory requirements.

In collaboration with Dr. Stacy Lindborg, Vice President, and Dr. Ying Zhu, Director, of Biostatistics at Biogen Idec., as well as Dr. Nathan Stein from The Wharton School at the University of Pennsylvania, and Joseph J. Lee from Harvard University’s Department of Statistics, Arman is currently helping to develop powerful multiple comparison procedures that satisfy FDA requirements for evaluating the effectiveness of pharmaceuticals for neurodegenerative diseases.  Clinical trials, such as these, are plagued with patient noncompliance of the assigned treatments due to ethical and safety standards regarding patient survival.  To address these issues, Dr. Sabbaghi is drawing on concepts from causal inference and Bayesian statistics to perform powerful and valid multiple comparisons for a specific subpopulation of subjects who will always comply with their assigned treatment.

Among his interest in engineering and multiple comparisons, Arman also looks to engage those from various fields of study including:

  • Education- designing experiments or observational studies that evaluate the effectiveness of new educational techniques,
  • Economics- analyzing observational studies that are complicated by issues such as interference among units, censoring/truncation ‘due to death’, or missing data,
  • Biomedical Engineering- designing experiments to evaluate the effectiveness of new medical devices.

Dr. Arman Sabbaghi’s research program is focused on all aspects of experimental design and analysis.  He welcomes new Purdue collaborations and can be reached via email: sabbaghi@purdue.edu.