Category Archives: Seminars

Department of Statistics, Seminars

Statistics Seminar

“Recent Advances on Optimal Design of Experiments”
Dr. Min Yang, University of Illinois, Chicago

Date: Thursday, April 16, 2015
Time: 3:30 PM – 4:30 PM
Location: 160 English Building
NOTE: This is different than our usual location.
Sponsor: Department of Statistics, Annie Qu

Abstract:
Designing experiments is an integral part of the scientific process, both for discovery and verification. Resources are almost always scarce, and judicious use of the limited resources is essential. Identifying efficient and optimal designs for data collection is therefore paramount. In this talk, I will first give one simple example demonstrating the set up and the impact of efficient designs. Then I will review the classical tools of studying optimal designs. A new tool will be introduced. With this tool, many well-known results in the optimal design literature can be unified and extended. If time permitted, I plan to talk about a newly developed algorithm. The new algorithm can be applied to variety of optimality problems. More important it is much faster than the existing algorithms.

http://illinois.edu/calendar/detail/1439?eventId=32137177&calMin=201504&cal=20150413&skinId=13335

Department of Statistics, Seminars

Statistics Seminar

Statistics Seminar

“Prior-free Probabilistic Inference: Inferential Models”
Dr. Chuanhai Liu, Purdue University

Date: Thursday, March 12, 2015
Time: 3:30 PM – 4:30 PM
Location: 165 Everitt
Sponsor: Department of Statistics; Annie Qu, and Bo Li

Abstract: Developing solid foundations for scientific inference is the most fundamental but unsolved problem in statistics. We argue for two basic principles for truly prior-free probabilistic inference and introduce a new framework, which is built upon the two principles and called Inferential Models (IMs). It follows as a remark that IM bridges the gap between the Bayesian and frequentist schools of thought that has been previously considered impossible. In this talk, we introduce IMs with simple illustrative examples.

* This is joint work with Ryan Martin at UIC.

http://illinois.edu/calendar/detail/1439?eventId=32297522&calMin=201503&cal=20150309&skinId=1

Department of Statistics, Seminars

Upcoming Statistics Seminar – Tuesday, March 10, 2015

Joint Mathematics and Statistics Colloquium

“The Yule-Coalescent Hierarchical Model”
Dr. James Degnan, University of New Mexico

Date: Tuesday, March 10, 2015
Time: 4:00 PM – 5:00 PM
Location: 165 Everitt
Sponsor: Departments of Mathematics and Statistics

Abstract:
Evolutionary biologists work with data at multiple levels which can be integrated into a hierarchical model: branching processes to model populations splitting (speciation), genes forming trees of ancestry within these populations, and mutations in genetic lineages.  Although much work has been done on each of these levels separately, there is much to understand regarding how the levels interact. By averaging effects at intermediate levels, we can study impacts of higher level processes, such as speciation and extinction rates, on lower levels, such as patterns in gene trees and DNA sequences, and in turn estimate parameters from higher levels using data collected at the lower levels.  In this talk I will examine some consequences of the Yule branching process model of speciation on distances between evolutionary trees and measures of tree balance.

http://illinois.edu/calendar/detail/1439?eventId=32516970&calMin=201503&cal=20150305&skinId=1

Career opportunities, Informational Meeting, Seminars

March 3: Government Webinar

 
Go Government Webinar: 
Interviews and Security Clearances: What to Expect
 
Date: Tuesday, March 3
Time: Noon-1:00pm
Location: The Career Center Conference Room
 
So you’ve applied for a federal job: now what?  Join us as we discuss how to prepare for a federal interview and what to expect from the security clearance process.
 
The Partnership for Public Service’s Go Government webinar series are designed for career services professionals to share with students and alumni looking for federal opportunities.
 
Please feel free to contact me at vspring@illinois.edu if you have any questions.

 

Department of Statistics, Seminars, Undergraduate research

Rose-Hulman Annual Undergraduate Mathematics Conference

Hi Mathematics and Statistics colleagues!
 
The Rose-Hulman Mathematics Department is excited to announce that the registration for the 2015 Rose-Hulman Undergraduate Mathematics Conference on April 24-25, 2015 is now open. This will be Rose-Hulman’s 32nd consecutive year hosting this conference!
 
As in past years, the primary vision for the conference is to provide a venue for highlighting and celebrating the accomplishments and work of undergraduate students in mathematics and statistics. Last year, the conference attracted more than 125 students and faculty, with 25 students presenting their work. This year’s conference theme is Statistics: From Big Data to Big Decisions, and we will be showcasing the field of statistics and its career opportunities. As plenary speakers, we have chosen two individuals who use statistics in their daily work. The first speaker, Nick Lockwood, is Senior Director of Brand Marketing Analytics at Catalina Marketing, a personalized digital media company with the largest shopper purchase history database in the world. The second speaker, Amanda Cox, is a graphics editor at the New York Times, where she makes charts and maps for the paper and its web site.
 
The student talks are the focal point of this conference, and we encourage you to seek out those students at your school who’d be enthusiastic about presenting their work. There is no registration fee for students at this year’s conference, and we will be able to offer discounts on hotel arrangements.  We’ve been working with publishers to be able to provide current and classic texts in mathematics and statistics to award to student speakers. Also, this year we are sponsoring the 2015 Distinguished Paper Award, which will be presented to a student presenting a paper which, in the minds of the award selection committee, is an exceptional example of undergraduate work in a mathematical discipline. The recipient of the award will receive a $100 cash prize in addition to a certificate.
 
All of this information with more details is available at the conference website: http://www.rose-hulman.edu/mathconf. Faculty and students can register at the website, and the site has important key dates about registration and paper submission. The deadline for submitting a paper for the Distinguished Paper Award is March 16, and the deadline to submit a talk is April 2.
 
We are looking forward to seeing some familiar faces along with some new ones at this year’s conference.  We hope that our Terre Haute neighbors at Indiana State University and St. Mary-of-the-Woods College, as well as local high school students and teachers, will be able to join us this year. We are delighted to have our friends up north from Purdue attending, and we’d really like to have other schools in Indiana and surrounding states join us!  We certainly welcome your questions about the conference if you can’t find the information that you need at the website – please email either one of us!
Thank you in advance for sharing this email with your students and encouraging them to give talks!
 
Dr. Diane Evans (diane.evans@rose-hulman.edu)
2015 Conference Co-organizer
 
Dr. Eric Reyes (reyesem@rose-hulman.edu)
2015 Conference Co-organizer
 
Rose-Hulman Institute of Technology
Mathematics Department
5500 Wabash Avenue
Terre Haute, IN 47803

 

Activities, Department of Statistics, Seminars

SAS Day – February 11

SAS Day is a nice opportunity to learn about statistical applications in SAS and about work current student interns and alumni are doing at State Farm. A list of speakers and presentations are available through the link below. The event is free and advance registration is available through the Webstore.

Registration is open for SAS Day.  It will be on February 11th at the Union.  We have an exciting agenda planned. Here is the link to the Registration page:

https://webstore.illinois.edu/home/sasday.aspx

Department of Statistics, Seminars

Statistics Seminar – Tomorrow – Tuesday December 2nd

“An L2 Test Theory for Nonstationary Time Series”
Dr. Wei Biao Wu, University of Chicago

Date: Tuesday, December 02, 2014
Time: 4:00 PM – 5:00 PM
Location: 269 Everitt
Sponsor: Department of Statistics, Xiaohui Chen

Abstract:
I will discuss testing whether the mean trend of a nonstationary time series is of certain parametric forms. A central limit theorem for the integrated squared error is derived, and with that a hypothesis-testing procedure is proposed. The method is illustrated in a simulation study, and is applied to assess the trend pattern in the central England temperature series. The work is joint with Ting Zhang.

Department of Statistics, Seminars

Statistics Seminar on December 2nd

“An L2 Test Theory for Nonstationary Time Series”
Dr. Wei Biao Wu, University of Chicago

Date: Tuesday, December 02, 2014
Time: 4:00 PM – 5:00 PM
Location: 269 Everitt
Sponsor: Department of Statistics, Xiaohui Chen

Abstract:
I will discuss testing whether the mean trend of a nonstationary time series is of certain parametric forms. A central limit theorem for the integrated squared error is derived, and with that a hypothesis-testing procedure is proposed. The method is illustrated in a simulation study, and is applied to assess the trend pattern in the central England temperature series. The work is joint with Ting Zhang.

Department of Statistics, Seminars

Statistics Seminar


“L2 Asymptotics for High-Dimensional Data”
Dr. Mengyu Xu, University of Chicago

Date: Thursday, November 20, 2014
Time: 4:00 PM – 5:00 PM
Location: 269 Everitt
Sponsor: Department of Statistics, Xiaohui Chen

Abstract:

We develop an asymptotic theory for L2 norms of sample mean vectors of high-dimensional data. An invariance principle for the L2 norms is derived under conditions that involve a delicate interplay between the dimension p, the sample size n and the moment condition. Under proper normalization, central and non-central limit theorems are obtained. To facilitate the related statistical inference, we propose a plug-in calibration method and a sub-sampling procedure to approximate the distributions of the L2 norms. Our results are applied to multiple tests and inference of covariance matrix structures.