SED navigation bar go to SED home page go to SED projects page go to NIST home page SED Home Page SED Contacts SED Projects SED Products and Publications Search SED Pages

Education and Training: Combining Information

Time and Location Combining Information
James Yen
Statistical Engineering Division, NIST
Thursday, Friday September 26, 27, 2002 1:30pm - 4:30pm
Adminstration Building, Lecture Room D
Gaithersburg, MD
Abstract How best to combine information from different experiments or sources remains an area of continuing study. This 6 hour (two 3-hour afternoon sessions) workshop explores the statistical aggregation of information for three different kinds of problems:
  1. Combining information for parameter estimation. This includes analysis of weighted means and combining results from multiple methods, along with the relevant uncertainties.

  2. Combining information for decision making. Topics include combined hypothesis tests, simultaneous statistical inference and confidence intervals, effect sizes, and combined classification procedures.

  3. Combining information for function estimation. Estimating a "consensus function" and relevant uncertainties can be quite problematic. Here, we present several approaches including local regression (weighted and otherwise).
In all 3 sections, the techniques and analyses will be demonstrated on data sets encountered at NIST.

This workshop has about 30% overlap with last year's meta-analysis course; it will be more of a special topics course rather than a unified treatment, and there will be a greater emphasis on examples rather than theory. Students would ideally have had a previous course in introductory statistics.

Comments on Course CLASS SIZE IS LIMITED TO 40.

A set of notes will be provided for the class. There is no textbook.

Further Information For further information, contact or register online at

Date created: 9/10/2002
Last updated: 9/10/2002
Please email comments on this WWW page to

SED Home |  Education Home |  Previous ] Next ]