Abstract #302064

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.


Back to main JSM 2003 Program page



JSM 2003 Abstract #302064
Activity Number: 446
Type: Contributed
Date/Time: Thursday, August 7, 2003 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Computing
Abstract - #302064
Title: Discovering ANOVA Structure in Black Box Functions
Author(s): Giles Hooker*+
Companies: Standford University
Address: 340 Marmona Dr., Menlo Park, CA, 94025,
Keywords: functional ANOVA ; data mining ; diagnostics ; interpretation ; machine learning
Abstract:

Many automated learning proceedures lack interpretability, often operating effectively as a black box--providing a prediction tool, but no explanation of the underlying dynamics that drive it. A common approach to interpretation is to plot the dependence of a learned function on one or two predictors. I present a method that seeks not to display behavior of a function, but to evaluate the importance of functional ANOVA interactions within any set of variables. This is turned into an algorithm to determine what components are needed to sucessfully reproduce the function as an additive model. An immediate application is a diagnostic for how much information low dimensional plots provide about function behavior. The calculations used here correspond closely with the functional ANOVA decomposition. In particular, the measure of importance used can be expressed as a sum of variances of ANOVA components, and measures the loss associated with the projection of the function onto a space of additive models. The algorithm runs in feasible time and I present displays of the output as a graphical model of the function for interpretation purposes.


  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003