Published 2001 | Version v1
Miscellaneous

Identifying multiple outliers in linear regression: robust fit and clustering approach

  • 1. Universiti Teknologi Malaysia, Skudai (Malaysia). Faculty of Science
  • 2. Universiti Teknologi Malaysia, Skudai (Malaysia). Faculty of Geoinformation Science and Engineering

Description

This research provides a clustering based approach for determining potential candidates for outliers. This is modification of the method proposed by Serbert et. al (1988). It is based on using the single linkage clustering algorithm to group the standardized predicted and residual values of data set fit by least trimmed of squares (LTS). (Author)

Part of:
Proceedings of the Malaysian Science and Technology Congress 2000: Symposium C, Volume VI

Additional details

Publishing Information

Publisher
Confederation of Scientific and Technological Associations in Malaysia COSTAM
Imprint Place
Petaling Jaya (Malaysia)
Imprint Title
Proceedings of the Malaysian Science and Technology Congress 2000: Symposium C, Vol. VI
Imprint Pagination
400 p.
Journal Page Range
p. 226-233

Conference

Title
Research and Development in Science and Technology for the New Era - Symposium C
Acronym
Malaysian Science and Technology Congress 2000
Dates
7-9 Nov 2000
Place
Genting Highlands (Malaysia)

INIS

Country of Publication
Malaysia
Country of Input or Organization
Malaysia
INIS RN
34031908
Subject category
S99: GENERAL AND MISCELLANEOUS;
Resource subtype / Literary indicator
Conference, Non-conventional Literature
Descriptors DEI
ALGORITHMS; CLUSTER MODEL; LEAST SQUARE FIT; MODIFICATIONS; REGRESSION ANALYSIS
Descriptors DEC
MATHEMATICAL LOGIC; MATHEMATICAL MODELS; MATHEMATICS; MAXIMUM-LIKELIHOOD FIT; NUCLEAR MODELS; NUMERICAL SOLUTION; STATISTICS

Optional Information