| 
|
Readme file
SERIES
B
 Statistical
Methodology
B. Li: A new approach to cluster analysis: the clustering-function-based method
Journal of the Royal Statistical Society, Series B, volume
68 (2006), part 3, pages 457 - 476
The data files consist of five documents, ‘dataExample3.txt’, ‘dataExample4.txt’, ‘clusteringfunction.m’, ‘clusteringfunction.p’,
and ‘helpDocument.pdf’.
The file ‘dataExample3.txt’ contains a simulated data matrix
of size 300 by 20 used in example 3 in the paper, representing 300 objects
measured on 20 variables. The first 100 objects were generated from population
1, the second 100 objects were generated from population 2 and the last
100 objects were generated from population 3.
The file ‘dataExample4.txt’ contains a simulated data matrix
of size 200 by 3 used in example 4 in the paper, representing 200 objects
measured on 3 variables. The first 80 objects were generated from population
1, and the remaining 120 objects were generated from population 2.
The software package containing a Matlab routine can be found in the file ‘clusteringfunction.m’.
The primary function is ‘clusteringfunction.m’, followed by
several subfunctions. The file ‘clusteringfunction.p’ is the
p-code version of ‘clusteringfunction.m’. Either ‘clusteringfunction.m’ or ‘clusteringfunction.p’ can
be used to perform clustering-function-based analysis. Please refer to ‘clusteringfunction.m’ if
you are interested in the algorithms and pseudocode; otherwise simple call ‘clusteringfunction.p’ for
applications.
The help information about ‘clusteringfunction.m’ is provided
in the document ‘helpDocument.pdf’.
For more information, please contact:
Baibing Li
Business School
Loughborough University
Ashby Road
Loughborough, Leicestershire
LE11 3TU
UK
E-mail: b.li2@lboro.ac.uk
|
Journals
SERIES
A
Statistics
in Society
SERIES
B
Statistical
Methodology
SERIES C
Applied Statistics
SERIES D
The
Statistician
|
