Emergency BU Alert Boston University's Charles River and Medical Center Campuses will be closed all day Tuesday, January 27, 2015. When classes resume, they will resume on the regular class schedule. Whether or how classes are to be made up is at the discretion of the faculty member. Please note: Employees in essential services must report as scheduled. Essential services include, but are not limited to, University Police, Public Safety, Emergency Patient Treatment, Facilities Management and Planning, Environmental Health & Safety, University Dining Services, Mail Services, Student Health Services, Occupational and Environmental Medicine and the University Switchboard. For the very latest information, please go to BU Today at http://www.bu.edu/today and the Emergency Communications page at http://www.bu.edu/ehs/comm

Impactx2 Logo and Photo

Impact x2 Qais

Impactx2 Content

How can we work together to promote better cultural understanding worldwide?

Qais Akbar Omar (GRS’16), a graduate student in the Creative Writing Program, has published a much-praised memoir, A Fort of Nine Towers: An Afghan Family Story. He recalls how the violence and tumult of civil war jolted his family, who, despite losing relatives, their home, and possessions, continued to nurture his wish to attend a university.

Impactx2 Call to Action

With your help, students like Qais gain the skills they need to tell their story and give us a broader understanding of the world.

Will you support CAS?

Samuel Kou - Harvard University

Starts:
4:00 pm on Thursday, November 15, 2012
Ends:
5:00 pm on Thursday, November 15, 2012
Location:
MCS 148
Title: Optimal Shrinkage Estimation in Heteroscedastic Hierarchical Models. Abstract: Hierarchical models are powerful statistical tools widely used in scientific and engineering applications. The homoscedastic (equal variance) case has been extensively studied, and it is well known that shrinkage estimates, the James-Stein estimate in particular, have nice theoretical (e.g., risk) properties. The heteroscedastic (the unequal variance) case, on the other hand, has received less attention, even though it frequently appears in real applications. It is not clear of how to construct "optimal" shrinkage estimate. In this talk, we study this problem. We introduce a class of shrinkage estimates, inspired by Stein's unbiased risk estimate. We will show that this class is asymptotically optimal in the heteroscedastic case. We apply the estimates to real examples and observe excellent numerical results. This talk is based on joint work with Lawrence Brown and Xianchao Xie.