Novel Approaches to Identify Clusters Using Independent Components Analysis with Application

Table 3

The ICA-based clusters using ranked approach (case: JADE).

Cluster no.	Cluster size	Companies
1	34	ACPL, ALNRS, BAHL, BNWM, BWHL, CHCC, DAWH, DGKC, DNCC, DSFL, EFUG, ESBL, FABL, GLAXO, GTYR, HINO, HINOON, HUBC, IDRT, JDWS, KESC, KOIL, LUCK, MARI, MFTM, MIRKS, MLCF, MODAM, MUKT, OGDC, PAEL, PIAA, SAIF, and SCM
2	12	ATBA, BGL, GADT, GLPL, JOPP, JSCL, KASBB, NESTLE, OLPL, PAKMI, PAKT, and PSO
3	11	AGIC, CPL, FPJM, IBFL, ICI, IDYM, PAKRI, PKGS, RICL, SAZEW, and SEPCO
4	9	ATLH, CLOV, EFUL, FDIBL, FUDLM, HICL, MUREB, NATF, and SING
5	6	AGIL, FCCL, GHNL, KOHE, PIOC, and SHFA
6	6	FHAM, KSBP, NCL, SHEL, SIEM, and SITC
7	5	GWLC, HSPI, PAKD, PCAL, and SIBL
8	4	ANL, FECTC, FNBM, and PRL
9	3	ABOT, BYCO, and MZSM
10	3	ADOS, ATRL, and ECOP
11	3	BIFO, MCB, and SANSM
12	3	CPMFI, FEROZ, and FHBM
13	3	ENGRO, FFC, and PNGRS
14	3	GASF, PNSC, and SEARL
15	2	ADMM, and CEPB
16	2	AICL and FFLM
17	2	CENI and FRCL
18	2	CHAS and CRTM
19	2	CSAP and DFML
20	2	FECM and PMI
21	2	FEM and NIB
22	2	PICT and RAVT