Accession Number:

ADA256106

Title:

Goals Versus Algorithms

Descriptive Note:

Final rept. 1 Jul 1989-30 Jun 1992

Corporate Author:

MASSACHUSETTS INST OF TECH CAMBRIDGE DEPT OF MATHEMATICS

Personal Author(s):

Report Date:

1992-06-30

Pagination or Media Count:

13.0

Abstract:

Computational methods in statistics often are defined through an algorithm. It is argued that a precise finite sample specification of the goals to be achieved by the algorithm is at least equally important. The issues are discussed in the special case of Projection Pursuit Regression. An interesting initial result of this work, which is still in progress, is that the Friedman- Stuetzle algorithm appears to be systematically biased toward overfitting.

Subject Categories:

  • Statistics and Probability

Distribution Statement:

APPROVED FOR PUBLIC RELEASE