|
People of the Methodology Center |
|
Home  People Eric B. Laber, M.A.
Eric B. Laber, M.A.
Research Assistant, The Quantitative Methodology Program
 |
Department of Statistics
439 West Hall, 1085 S. Univ.
University of Michigan
Ann Arbor, MI 48109-1107
|
 |
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
|
|
734-615-4607 |
|
| Education: |
|
Ph.D., University of Michigan, Expected 2011 (Statistics)
M.A., University of Michigan, 2007 (Statistics)
B.S., University of California at Los Angeles, 2005 (Mathematics and Computation)
|
| Research Interests: |
|
Statistical Inference in Machine Learning including small sample prediction intervals in RL and classification problems, the bootstrap in non-regular estimation problems arising from non-smooth criterion functions used to evaluate/train learners, and extrapolation of classifier performance to shifted validation populations.
Courtroom dynamics including the study of judge-prosecutor-defendant interactions and heterogeneous sentencing outcomes.
|
| Current Projects and Collaborators: |
|
I work with my advisor Susan A. Murphy on the problem of statistical inference in the context of dynamic treatment regimes. Specifically, our aim is in constructing valid prediction intervals for the cumulative effect of a sequence of treatments.
I also work with JJ Prescott of the University of Michigan Law School, Kathy Barnes and Mark Miller of the University of Arizona Law School, Ronald Wright of Wake Forest University on the problem of heterogeneous treatments of defendants across race, age, gender and other defendant demographics. I am also involved in a research project with Kerby Shedden of University of Michigan Statistics Department that involves extrapolating the performance of a learned classifier to (1) larger sample sizes, (2) a shifted validation population, or (3) both.
|
Honors and Awards:
|
|
|
|
|
