Prof. Paul Feigin

Statistics

Service Enterprise Engineering Laboratory, Statistics Laboratory

Professor Paul Feigin's research covers three main areas: inference for stochastic processes (such as for time series or discrete event sequences); modern methods of forecasting and data-mining, in which he has created models that have been used to forecast electricity demand; and design and analysis of industrial experiments and clinical trials. His most recent research projects include statistical analysis of genetic association studies and analysis of customer patience in call centers.


Professor Feigin has developed and implemented algorithms for building nonparametric time series models using both smoothing spline techniques as well as neural networks. These models have been successfully used to forecast electricity demand in Israel and abroad. A related approach has been used on a project for the Ministry of Energy to evaluate the effect of daylight savings time on electricity consumption. In the area of industrial experimentation and analysis, he is an expert on designing and analyzing Taguchi-type experiments. These experiments are designed to evaluate which settings are optimal both from the point of view of achieving design specifications on average, as well as from the point of view of minimizing variability that may arise from production variability, or from component deviations from nominal values. In the area of clinical trials, Professor Feigin has been involved in the statistical design of large-scale clinical trials and is a regular consultant for Teva Pharmaceuticals. 

Professor Feigin, together with Professor Ayala Cohen, heads the Technion Statistical Laboratory which provides a broad range of consulting services to academic researchers as well as to government and industry. These projects often involve the application of experimental design and analysis as well as sophisticated multivariate analysis techniques. Professor Feigin is a past president of the Israel Statistical Association, is an elected fellow of the Institute of Mathematical Statistics and of the International Statistical Institute.

Since 2013 Prof. Paul Feigin holds the position of the Technion Vice President for Strategic Projects.

 

Overview

Paul Feigin was appointed in 2013 as the Vice President for Strategic Projects, a newly-created position in which he is responsible for the planning and oversight of Technion's strategic projects overseas – in particular the China project, with partners Shantou University and the Li Ka Shing Foundation. From 2007, he served as Senior Executive Vice President, helping to forge the Technion’s partnership with Cornell University and its winning bid to build what is now the Joan & Irwin Jacobs Technion-Cornell Innovation Institute in New York City. A member of the academic faculty as well as the Senior Administration, he also served as Dean of The William Davidson Faculty of Industrial Engineering and Management from 1999–2002.

Holder of the Gruenblat Chair in Production Engineering, Prof. Feigin is an authority on statistical design and analysis of genomic and clinical studies. His research covers three main areas: inference for stochastic processes (such as for time series or discrete event sequences); modern methods of forecasting and data-mining, in which he has created models that have been used to forecast electricity demand; and design and analysis of industrial experiments. His most recent research projects include statistical analysis of genetic association studies and analysis of customer patience in call centers.

Prof. Feigin has extensive experience in industry. He has worked for Teva Pharmaceuticals as a statistical consultant, and until recently he served as the scientific director of TechnoSTAT—a data management and biostatistics company that provides services to the pharmaceutical and biotechnology industries. Together with Prof. Ayala Cohen, he headed the Technion Statistical Laboratory, which provides a broad range of consulting services to academic researchers as well as to government and industry.

Prof. Feigin earned his bachelor’s degree at the University of Melbourne in 1972 and his doctorate at The Australian National University in 1975 — both in statistics. He started his career at the Technion, joining the faculty in 1976.  He spent the academic year 1981-1982 as a visiting professor at the University of California, Berkeley and 1987-1988 as a visiting scientist in the Division of Mathematics and Statistics of the Commonwealth Scientific and Industrial Research Organization (CSIRO) in Melbourne, Australia. He has also been a visiting professor at Stanford University for several summer sessions, and has held short-term visiting positions at the University of Melbourne, Cornell University, The Wharton School at University of Pennsylvania and Monash University in Victoria, Australia.

He has published over 50 scientific papers, is a past president of the Israel Statistical Association, and is an elected fellow of the Institute of Mathematical Statistics and of the International Statistical Institute.  

Selected Publications

Azriel, David, Feigin, Paul D. Adaptive designs to maximize power in clinical trials with multiple treatments, Sequential Analysis, 33 (1), 60–86, 2014.

Azriel, David, Feigin, Paul D., Mandelbaum, Avishai.Erlang-S: A data-based model of servers in queueing networks, 2014.

Olanow, C Warren, Rascol, Olivier, Hauser, Robert, Feigin, Paul D., Jankovic, Joseph, Lang, Anthony, Langston, William, Melamed, Eldad, Poewe, Werner, Stocchi, Fabrizio.A double-blind, delayed-start trial of Rasagiline in Parkinson's disease, New England Journal of Medicine, 361 (13), 1268–1278, 2009.

Aldor-Noiman, Sivan, Feigin, Paul D., Mandelbaum, Avishai. Workload forecasting for a call center: Methodology and a case study, The Annals of Applied Statistics, 1403–1447, 2009.

Rosenberg, Shai, Templeton, Alan R., Feigin, Paul D., Lancet, Doron, Beckmann, Jacques S., Selig, Sara, Hamer, Dean H., Skorecki, Karl. The association of DNA sequence variation at the MAOA genetic locus with quantitative behavioural traits in normal males, Human Genetics, 120 (4), 447–459, 2006.

Feigin, Paul D., Resnick, Sidney I. Pitfalls of fitting autoregressive models for heavy-tailed time series, Extremes, 1 (4), 391–422, 1999.

Feigin, Paul D., Resnick, Sidney I. Linear programming estimators and bootstrapping for heavy tailed phenomena, Advances in Applied Probability, 759–805, 1997.

Feigin, Paul D., Resnick, Sidney I. Limit distributions for linear programming time series estimators, Stochastic Processes and their Applications, 51 (1), 135–165, 1994.

Feigin, Paul D., Resnick, Sidney I. Estimation for autoregressive processes with positive innovations, Communications in Statistics. Stochastic Models, 8 (3), 685–717, 1992.

Feigin, Paul D., Tweedie, Richard L. Linear functionals and Markov chains associated with Dirichlet processes, Mathematical Proceedings of the Cambridge Philosophical Society, 105 (3), 579–585, 1989, Cambridge University Press.

Feigin, Paul D., Alvo, Mayer. Intergroup diversity and concordance for ranking data: An approach via metrics for permutations, The Annals of Statistics, 691–707, 1986.

Feigin, Paul D., Tweedie, Richard L. Random coefficient autoregressive processes: a Markov chain analysis of stationarity and finiteness of moments, Journal of Time Series Analysis, 6 (1), 1–14, 1985.

Feigin, Paul D. Stable convergence of semimartingales, Stochastic Processes and Their Applications, 19 (1), 125–134, 1985.

Adler, Robert J., Feigin, Paul D. On the cadlaguity of random measures, The Annals of Probability, 615–630, 1984.

Alvo, Mayer, Cabilio, Paul, Feigin, Paul D. Asymptotic theory for measures of concordance with special reference to average Kendall tau, The Annals of Statistics, 1269–1276, 1982.

Feigin, Paul D. Conditional exponential families and a representation theorem for asymptotic inference, The Annals of Statistics,597–603, 1981.

Feigin, Paul D. On the characterization of point processes with the order statistic property, Journal of Applied Probability,297–304, 1979.

Feigin, Paul D. Some comments concerning a curious singularity, Journal of Applied Probability, 440–444, 1979.

Feigin, Paul D., Cohen, Ayala. On a model for concordance between judges, Journal of the Royal Statistical Society. Series B (Methodological), 203–213, 1978.

Feigin, Paul D. The efficiency criteria problem for stochastic processes, Stochastic Processes and their Applications, 6 (2), 115–127, 1978.

Feigin, Paul David. Maximum likelihood estimation for continuous-time stochastic processes, Advances in Applied Probability, 712–736, 1976.

Basawa, IV., Feigin, PD., Heyde, CC. Asymptotic properties of maximum likelihood estimators for stochastic processes, Sankhyā: The Indian Journal of Statistics, Series A, 259–270, 1976.

Feigin, PD., Heathcote, CR. The empirical characteristic function and the Cramer-von Mises statistic, Sankhyā: The Indian Journal of Statistics, Series A, 309–325, 1976.

Heyde, CC., Feigin, PD. On efficiency and exponential families in stochastic process estimation, A Modern Course on Statistical Distributions in Scientific Work, 227–240, 1975, Springer.

Research

Statistical design and analysis of genomic studies, analysis of Call Center data, data mining methods, design and analysis of clinical studies.
Methodology and implementation of data mining techniques.
Statistical genetics.
Topics in bio-statistics (especially related to clinical trials).
Service Enterprise Engineering Laboratory, Statistics Laboratory

Contact Info

Room 522 Bloomfield Building
+972-4-829-4508