DIMACS Center, Rutgers University, Piscataway, NJ

**Organizers:****Alan Karr**, National Institute of Statistical Sciences (NISS), karr@niss.org**Regina Liu**, Rutgers University, rliu@stat.rutgers.edu**David Johnson**, A T & T Research, dsj@research.att.com**Catherine McGeoch**, Amherst College, ccm@cs.amherst.edu**Joseph Naus**, Rutgers University, naus@stat.rutgers.edu**Fred Roberts**, DIMACS, froberts@dimacs.rutgers.edu

Title: A Brief Overview of Some Statistical Research Related to Computer Models

I'll summarize statistical ideas and methods that have been developed for empirical studies involving computer models. The presentation will focus on three general areas: (1.) assessment of the relative importance of inputs, (2.) quantitative understanding of input-output relationships, and (3.) questions involving relationships between models and their context. Most of the specific work to which I'll refer is motivated by problems involving computer models constructed to represent a ``reality'' of some sort; I'll conclude with some thoughts concerning how these and other statistical ideas might be useful in evaluations of more general models and algorithms.

Previous: Program

Workshop Index

DIMACS Homepage

Contacting the Center

Document last modified on March 29, 2006.