Experimental Design for (Semi-)Local Regression

We discuss a new problem statement of experimental design, which is applicable when a regression model is not completely specified. The proposed approach is a local regression model. The notion of local design extendability is introduced and basic properties of extendable designs are established. Th...

Full description

Saved in:
Bibliographic Details
Published in:Communications in statistics. Theory and methods 2003-01, Vol.32 (5), p.1035-1055
Main Authors: Rafajłowicz, E., Schwabe, R.
Format: Article
Language:English
Subjects:
Exact sciences and technology
Experimental design
Extendability
Linear inference, regression
Local regression
Mathematics
Probability and statistics
Sciences and techniques of general use
Statistics
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We discuss a new problem statement of experimental design, which is applicable when a regression model is not completely specified. The proposed approach is a local regression model. The notion of local design extendability is introduced and basic properties of extendable designs are established. Those extendable designs allow for an exploratory analysis of adjacent domains, retaining part of the observations gained earlier. Under certain conditions locally D-optimal designs can be extended to global optimal solutions.
ISSN:0361-0926
1532-415X
DOI:10.1081/STA-120019961