Accession Number:

ADP013738

Title:

Nonparametric Regression Subject to a Given Number of Local Extreme Value

Descriptive Note:

Conference paper

Corporate Author:

ESSEN UNIV (GERMANY) DEPT OF MATHEMATICS AND COMPUTER SCIENCE

Personal Author(s):

• Majidi, Ali
• Davis, Laurie

Report Date:

2001-07-01

Pagination or Media Count:

8.0

Abstract:

We consider the problem of nonparametric regression. The aim is to get a smooth function which represents the dataset and has a reasonable number of extreme values. An iterative method, the QSOR method is introduced. Problems with the slow convergence of the method are reduced using multigrid techniques.

Descriptors:

• *NONPARAMETRIC STATISTICS
• *REGRESSION ANALYSIS
• SYMPOSIA
• WHITE NOISE
• QUADRATIC PROGRAMMING
• GERMANY
• ITERATIONS
• FUNCTIONS(MATHEMATICS)
• SMOOTHING(MATHEMATICS)

Subject Categories:

• Numerical Mathematics
• Statistics and Probability

Distribution Statement:

APPROVED FOR PUBLIC RELEASE