AMSTATNEWS the Membership Magazine of the American Statistical Association •
Total Page:16
File Type:pdf, Size:1020Kb
Load more
Recommended publications
-
Higher-Order Asymptotics
Higher-Order Asymptotics Todd Kuffner Washington University in St. Louis WHOA-PSI 2016 1 / 113 First- and Higher-Order Asymptotics Classical Asymptotics in Statistics: available sample size n ! 1 First-Order Asymptotic Theory: asymptotic statements that are correct to order O(n−1=2) Higher-Order Asymptotics: refinements to first-order results 1st order 2nd order 3rd order kth order error O(n−1=2) O(n−1) O(n−3=2) O(n−k=2) or or or or o(1) o(n−1=2) o(n−1) o(n−(k−1)=2) Why would anyone care? deeper understanding more accurate inference compare different approaches (which agree to first order) 2 / 113 Points of Emphasis Convergence pointwise or uniform? Error absolute or relative? Deviation region moderate or large? 3 / 113 Common Goals Refinements for better small-sample performance Example Edgeworth expansion (absolute error) Example Barndorff-Nielsen’s R∗ Accurate Approximation Example saddlepoint methods (relative error) Example Laplace approximation Comparative Asymptotics Example probability matching priors Example conditional vs. unconditional frequentist inference Example comparing analytic and bootstrap procedures Deeper Understanding Example sources of inaccuracy in first-order theory Example nuisance parameter effects 4 / 113 Is this relevant for high-dimensional statistical models? The Classical asymptotic regime is when the parameter dimension p is fixed and the available sample size n ! 1. What if p < n or p is close to n? 1. Find a meaningful non-asymptotic analysis of the statistical procedure which works for any n or p (concentration inequalities) 2. Allow both n ! 1 and p ! 1. 5 / 113 Some First-Order Theory Univariate (classical) CLT: Assume X1;X2;::: are i.i.d. -
Thesis Is in Two Parts
Downloaded from orbit.dtu.dk on: Oct 03, 2021 Statistical Learning with Applications in Biology Nielsen, Agnes Martine Publication date: 2019 Document Version Publisher's PDF, also known as Version of record Link back to DTU Orbit Citation (APA): Nielsen, A. M. (2019). Statistical Learning with Applications in Biology. Technical University of Denmark. DTU Compute PHD-2018 Vol. 488 General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Users may download and print one copy of any publication from the public portal for the purpose of private study or research. You may not further distribute the material or use it for any profit-making activity or commercial gain You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Statistical Learning with Applications in Biology Agnes Martine Nielsen Kongens Lyngby 2018 PhD-2018-488 Technical University of Denmark Department of Applied Mathematics and Computer Science Richard Petersens Plads, building 324, 2800 Kongens Lyngby, Denmark Phone +45 4525 3031 [email protected] www.compute.dtu.dk PhD-2018-488 Summary (English) Statistical methods are often motivated by real problems. We consider methods inspired by problems in biology and medicine. The thesis is in two parts. -
IMS Bulletin July/August 2004
Volume 33 Issue 4 IMS Bulletin July/August 2004 A Message from the (new) President Louis H Y Chen, Director of the Institute CONTENTS for Mathematical Sciences at the 2-3 Members’ News; National University of Singapore, is the Contacting the IMS IMS President for 2004–05. He says: hen I was approached by the 4 Profi le: C F Jeff Wu WCommittee on Nominations in 5 IMS Election Results: January 2003 and asked if I would be President-Elect and Council willing to be a possible nominee for IMS 7 UK Research Assessment; President-Elect, I felt that it was a great Tweedie Travel Award honor for me. However, I could not help 8 Mini-meeting Reports but think that the outcome of the nomi- nation process would most likely be a 10 Project Euclid & Google nominee who is based in the US, because, 11 Calls Roundup except for Willem van Zwet, Nancy Reid of probabilists and statisticians. 14 IMS Meetings and Bernard Silverman, all the 68 past Although IMS is US-based, its infl u- Presidents of IMS were US-based. When ence goes far beyond the US due to its 20 Other Meetings and I was fi nally chosen as the nominee for several fi rst-rate publications and many Announcements President-Elect, I was pleased, not so high quality meetings. Also, IMS has 23 Employment much because I was chosen, but because I reduced membership dues for individuals Opportunities took it as a sign that the outlook of IMS in developing countries to encourage 25 International Calendar of was becoming more international. -
Raymond J. Carroll
Raymond J. Carroll Young Investigator Award Ceremony Wednesday, March 9, 2016 11:30 am – 12:30 pm Texas A&M Blocker Building Room 149 Daniella M. Witten 2015 Recipient Associate Professor of Biostatistics and Statistics University of Washington Raymond J. Carroll, Distinguished Professor The Raymond J. Carroll Young Investigator Award was established to honor Dr. Raymond J. Carroll, Distinguished Professor of Statistics, Nutrition and Toxicology, for his fundamental contributions in many areas of statistical methodology and practice, such as measurement error models, nonparametric and semiparametric regression, nutritional and genetic epidemiology. Carroll has been instrumental in mentoring and helping young researchers, including his own students and post-doctoral trainees, as well as others in the statistical community. Dr. Carroll is highly regarded as one of the world’s foremost experts on problems of measurement error, functional data analysis, semiparametric methods and more generally on statistical regression modeling. His work, characterized by a combination of deep theoretical effort, innovative methodological development and close contact with science, has impacted a broad variety of fields, including marine biology, laboratory assay methods, econometrics, epidemiology and molecular biology. In 2005, Raymond Carroll became the first statistician ever to receive the prestigious National Cancer Institute Method to Extend Research in Time (MERIT) Award for his pioneering efforts in nutritional epidemiology and biology and the resulting advances in human health. Less than five percent of all National Institutes of Health-funded investigators merit selection for the highly selective award, which includes up to 10 years of grant support. The Carroll Young Investigator Award is awarded biennially on odd numbered years to a statistician who has made important contributions to the area of statistics. -
BFF Workshop Participant List April 28 – May 1, 2019
BFF Workshop Participant List April 28 – May 1, 2019 Pierre Barbillon Gonzalo Garcia-Donato AgroParisTech Universidad de Castilla-La Mancha Samopriya Basu Edward George UNC - Chapel Hill Wharton, University of Pennsylvania Jim Berger Malay Ghosh Duke University University of Florida David R. Bickel Subhashis Ghoshal University of Ottawa North Carolina State University Alisa Bokulich Ruobin Gong Boston University Rutgers University Sudip Bose Mengyang Gu George Washington University Johns Hopkins University Fang Cai Yawen Guan Stanford University NC State University / SAMSI Hongyuan Cao Jan Hannig Florida State University UNC Chapel Hill Iain Carmichael Leah Henderson UNC Chapel Hill University of Groningen Jesse Clifton Wei Hu North Carolina State University Univeristy of California, Irvine Philip Dawid Michael Jordan University of Cambridge University of California, Berkeley David Dunson Kevin Kelly Duke University Carnegie Mellon University Anabel Forte-Deltell Todd Kuffner Universitat de Valencia Washington University, St. Louis Donald Fraser Subrata Kundu University of Toronto George Washington University BFF Workshop Participant List April 28 – May 1, 2019 Thomas Lee Shyamal Peddada University of California, Davis University of Pittsburgh Xinyi Li Elmor Peterson SAMSI Retired Gang Li Bruce Pitman University of North Carolina at Chapel University at Buffalo Hill Zhengling Qi Cong Lin University of North Carolina at Chapel East China Normal University Hill Regina Liu Nancy Reid Rutgers University University of Toronto Pulong Ma Ramchandra -
Trevor John Hastie 1040 Campus Drive Stanford, CA 94305 Home Phone&FAX: (650) 326-0854
Trevor John Hastie 1040 Campus Drive Stanford, CA 94305 Home Phone&FAX: (650) 326-0854 Department of Statistics Born: June 27, 1953, South Africa Sequoia Hall Married, two children Stanford University U. S. citizen, S.A. citizen Stanford, CA 94305 E-Mail: [email protected] (650) 725-2231 Fax: 650/725-8977 Updated: June 22, 2021 Present Position 2013{ John A. Overdeck Professor of Mathematical Sciences, Stanford University. 2006{2009 Chair, Department of Statistics, Stanford University. 2005{2006 Associate Chair, Department of Statistics, Stanford University. 1999{ Professor, Statistics and Biostatistics Departments, Stanford University. Founder and co-director of Statistics department industrial affiliates program. 1994{1998 Associate Professor (tenured), Statistics and Biostatistics Departments, Stan- ford University. Research interests include nonparametric regression models, computer in- tensive data analysis techniques, statistical computing and graphics, and statistical consulting. Currently working on adaptive modeling and predic- tion procedures, signal and image modeling, and problems in bioinformatics with many more variables than observations. Education 1984 Stanford University, Stanford, California { Ph.D, Department of Statis- tics (Werner Stuetzle, advisor) 1979 University of Cape Town, Cape Town, South Africa { First Class Masters Degree in Statistics (June Juritz, advisor). 1976 Rhodes University, Grahamstown, South Africa { Bachelor of Science Honors Degree in Statistics. 1975 Rhodes University, Grahamstown, South Africa { Bachelor of Science Degree (cum laude) in Statistics, Computer Science and Mathematics. Awards and Honors 2020 \Statistician of the year" award, Chicago chapter ASA. 2020 Breiman award (senior) 2019 Recipient of the Sigillum Magnum, University of Bologna, Italy. 2019 Elected to The Royal Netherlands Academy of Arts and Science. 2019 Wald lecturer, JSM Denver. -
Curriculum Vitae
Curriculum Vitae Nancy Margaret Reid O.C. April, 2021 BIOGRAPHICAL INFORMATION Personal University Address: Department of Statistical Sciences University of Toronto 700 University Avenue 9th floor Toronto, Ontario M5S 1X6 Telephone: (416) 978-5046 Degrees 1974 B.Math University of Waterloo 1976 M.Sc. University of British Columbia 1979 Ph.D. Stanford University Thesis: “Influence functions for censored data” Supervisor: R.G. Miller, Jr. 2015 D.Math. (Honoris Causa) University of Waterloo Employment 1-3/2020 Visiting Professor SMRI University of Sydney 1-3/2020 Visiting Professor Monash University Melbourne 10-11/2012 Visiting Professor Statistical Science University College, London 2007-2021 Canada Research Chair Statistics University of Toronto 2003- University Professor Statistics University of Toronto 1988- Professor Statistics University of Toronto 2002-2003 Visiting Professor Mathematics EPF, Lausanne 1997-2002 Chair Statistics University of Toronto 1987- Tenured Statistics University of Toronto 1986-88 Associate Professor Statistics University of Toronto Appointed to School of Graduate Studies 1986/01-06 Visiting Associate Professor Mathematics Univ. of Texas at Austin 1985/07-12 Visiting Associate Professor Biostatistics Harvard School of Public Health 1985-86 Associate Professor Statistics Univ. of British Columbia Tenured Statistics Univ. of British Columbia 1980-85 Assistant Professor Statistics & Mathematics Univ. of British Columbia 1979-80 Nato Postdoctoral Fellow Mathematics Imperial College, London 1 Honours 2020 Inaugural -
Big Data Fazekas 2020.Pdf
Using Big Data for social science research Mihály Fazekas School of Public Policy Central European University Winter semester 2019-20 (2 credits) Class times : Fridays on a bi-weekly bases (6 blocks) Office hours: By appointment (Quellenstrasse building) Teaching Assistant: Daniel Kovarek Version: 9/12/2019 Introduction The course is an introduction to state-of-the-art methods to use Big Data in social sciences research. It is a hands-on course requiring students to bring their own research problems and ideas for independent research. The course will review three main topics making Big Data research unique: 1. New and emerging data sources such social media or government administrative data; 2. Innovative data collection techniques such as web scraping; and 3. Data analysis techniques typical of Big Data analysis such as machine learning. Big Data means that both the speed and frequency of data created are increasing at an accelerating pace virtually covering the full spectrum of social life in ever greater detail. Moreover, much of this data is more and more readily available making real-time data analysis feasible. During the course students will acquaint themselves with different concepts, methodological approaches, and empirical results revolving around the use of Big Data in social sciences. As this domain of knowledge is rapidly evolving and already vast, the course can only engender basic literacy skills for understanding Big Data and its novel uses. Students will be encouraged to use acquired skills in their own research throughout the course and continue engaging with new methods. Learning outcomes Students will be acquainted with basic concepts and methods of Big Data and their use for social sciences research. -
New Applied Statistics Specialist Program
Awards: Faculty & Students Program News Congratulations to Alison Gibbs wins Dean’s Outstanding Tenure & Promotion GRADUATE STUDIES REPORT Don Fraser! Teaching Award by Sebastian Jaimungal, Associate Chair for Graduate Studies, Associate Professor, Dept. Statistics, U. Toronto Congratulations to Andrei Badescu who has been First off, I would like to thank Prof. Knight, my our faculty’s research problems, but are acces- Congratulations to Don Fraser For her clear leadership and achievements in manuscript Lessons from Medicine for the training granted tenure and was promoted to the rank of predecessor, for handling the reigns as sible to non-experts. This year we have for his recent appointment by teaching and the widespread enthusiasm for her of Statistical Consultants. She chairs the Statistical Associate Professor on July 1, 2011. Professor Graduate Chair so expertly, for his guidance introduced: Statistical Dependence: Copula the Governor General as an performance as an instructor. Dr. Gibbs is a lead- Education Committee for the SSC and attends Badescu is a Professor in Actuarial Science. He is an and for his assistance in bringing me up to Models and Beyond; Functional Data Analysis Officer of the Order of Canada. This honour was ing innovator of statistics education and several conferences on statistics education on an internationally renowned expert in ruin theory, par- speed—thanks Keith! Second, I would like to and Related Topics; Monte Carlo Estimation; bestowed upon Don for his contributions to the curriculum renewal in our Department and more annual basis. She was Guest editor for the ticularly for his work connecting risk processes with thank the staff: Andrea, Angela, Annette and Advanced Monte Carlo Methods and advancement of statistical sciences in Canada broadly the Faculty of Arts and Science at the Canadian Journal of Statistics, is an Associate stochastic fluid flows. -
By Alex Reinhart Yosihiko Ogata
STATISTICAL SCIENCE Volume 33, Number 3 August 2018 A Review of Self-Exciting Spatio-Temporal Point Processes and Their Applications .................................................................................Alex Reinhart 299 Comment on “A Review of Self-Exciting Spatiotemporal Point Process and Their Applications”byAlexReinhart.............................................Yosihiko Ogata 319 Comment on “A Review of Self-Exciting Spatio-Temporal Point Process and Their Applications”byAlexReinhart..........................................Jiancang Zhuang 323 Comment on “A Review of Self-Exciting Spatio-Temporal Point Processes and Their Applications”byAlexReinhart..................................Frederic Paik Schoenberg 325 Self-Exciting Point Processes: Infections and Implementations .............Sebastian Meyer 327 Rejoinder: A Review of Self-Exciting Spatio-Temporal Point Processes and Their Applications..................................................................Alex Reinhart 330 On the Relationship between the Theory of Cointegration and the Theory of Phase Synchronization..................Rainer Dahlhaus, István Z. Kiss and Jan C. Neddermeyer 334 Confidentiality and Differential Privacy in the Dissemination of Frequency Tables ......................Yosef Rinott, Christine M. O’Keefe, Natalie Shlomo and Chris Skinner 358 Piecewise Deterministic Markov Processes for Continuous-Time Monte Carlo .....................Paul Fearnhead, Joris Bierkens, Murray Pollock and Gareth O. Roberts 386 Fractionally Differenced Gegenbauer Processes -
Ali Shojaie, Ph.D
Ali Shojaie, Ph.D. University of Washington Department of Biostatistics Phone: (206) 616-5323 334 Hans Rosling Center Email: [email protected] 3980 15th Avenue NE, Box 351617 Homepage: faculty.washington.edu/ashojaie/ Seattle, WA 98195 Education • Iran University of Science & Technology, Tehran, Iran, B.Sc in Industrial and Systems Engineering, 09/1993-02/1998 • Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran, M.Sc in Industrial Engineering, 09/1998-02/2001 • Michigan State University, East Lansing, MI, M.S. Statistics, 01/2003-05/2005 • University of Michigan, Ann Arbor, MI, M.S. in Human Genetics, 09/2006-12/2009 • University of Michigan, Ann Arbor, MI, M.S. in Applied Mathematics, 09/2007-04/2010 • University of Michigan, Ann Arbor, MI, PhD in Statistics, 09/2005- 04/2010 Professional Positions • Postdoctoral Research Fellow, Dept. of Statistics, University of Michigan, 05/2010-06/2011 • Visiting Scholar, Statistical & Applied Mathematical Sciences Institute (SAMSI), 09/2010-05/2010 • Professor of Biostatistics and Adjunct Professor of Statistics, University of Washington, 07/2020- present – Associate Professor of Biostatistics and Adjunct Associate Professor of Statistics, University of Washington, 07/2016-06/2020 – Assistant Professor of Biostatistics and Adjunct Assistant Professor of Statistics, University of Washington, 07/2011-06/2016 • Associate Chair for Strategic Research Affairs, Department of Biostatistics, University of Washington, 09/2020-present • Founding Director, Summer Institute for Statistics in Big Data (SISBID), 05/2014-present • Affiliate Member, Center for Statistics in Social Sciences (CSSS), University of Washington, 03/2012- present • Affiliate Member, Biostatistics and Biomathematics Program, Fred Hutchinson Cancer Research Center (FHCRC), 09/2012-present • Affiliate Faculty, eSciences Institute, University of Washington, 03/2014-present • Affiliate Member, Algorithmic Foundations of Data Science (ADSI) Institute, University of Washington, 2018-present Ali Shojaie, Ph.D. -
Selective Inference for Hierarchical Clustering Arxiv:2012.02936V1
Selective Inference for Hierarchical Clustering Lucy L. Gaoy,∗ Jacob Bien◦, and Daniela Witten z y Department of Statistics and Actuarial Science, University of Waterloo ◦ Department of Data Sciences and Operations, University of Southern California z Departments of Statistics and Biostatistics, University of Washington September 23, 2021 Abstract Classical tests for a difference in means control the type I error rate when the groups are defined a priori. However, when the groups are instead defined via clus- tering, then applying a classical test yields an extremely inflated type I error rate. Notably, this problem persists even if two separate and independent data sets are used to define the groups and to test for a difference in their means. To address this problem, in this paper, we propose a selective inference approach to test for a differ- ence in means between two clusters. Our procedure controls the selective type I error rate by accounting for the fact that the choice of null hypothesis was made based on the data. We describe how to efficiently compute exact p-values for clusters obtained using agglomerative hierarchical clustering with many commonly-used linkages. We apply our method to simulated data and to single-cell RNA-sequencing data. Keywords: post-selection inference, hypothesis testing, difference in means, type I error arXiv:2012.02936v2 [stat.ME] 22 Sep 2021 ∗Corresponding author: [email protected] 1 1 Introduction Testing for a difference in means between groups is fundamental to answering research questions across virtually every scientific area. Classical tests control the type I error rate when the groups are defined a priori.