Where academic tradition
meets the exciting future

The Self-Organizing Map in Financial Benchmarking

Tomas Eklund, The Self-Organizing Map in Financial Benchmarking. TUCS Dissertations 56. Turku Centre for Computer Science, 2004.


Managers in today´s international companies are facing a very complex
competitive environment. In an increasingly global market, it has become
more and more important to recognize who your competitors are, what they
are doing, and how they are performing. This is the objective of
competitor analysis.

Financial benchmarking is an essential component of competitor analysis.
Financial benchmarking is performance benchmarking using financial
easures. However, while today´s tools are adequate for measuring
performance according to a few ratios for a small number of competitors,
increasing dimensionality and data present difficulties for managers.
Analyzing ratios that measure different aspects of performance is
especially difficult using today´s tools. A state of the art survey of
financial benchmarking methods in Finnish publicly-noted companies found
that few advanced, multiple ratio methods are currently used. The survey
found support for the need for new, complexity-reducing tools in
financial benchmarking.

In this dissertation, a model for financial benchmarking in the
international pulp and paper industry has been built using the self-
organizing map (SOM). Financial data for 98 companies in the
international pulp and paper industry have been collected for the years
1995-2002, and seven financial ratios measuring different aspects of
financial performance were calculated. A SOM model has been built, and
the benchmarking of the companies has been tested. The model has been
evaluated by subject matter experts from industry.

The study has found that the managers considered the model better than
many of their own methods, especially in terms of format. In particular,
the model was found to be useful in strategic decision making settings.
The results thus indicate that the SOM is a feasible tool for financial
benchmarking, and presents several advantages over current methods.


Full publication in PDF-format

BibTeX entry:

  title = {The Self-Organizing Map in Financial Benchmarking},
  author = {Eklund, Tomas},
  number = {56},
  series = {TUCS Dissertations},
  school = {Turku Centre for Computer Science},
  year = {2004},
  keywords = {Financial benchmarking, state of the art, pulp and paper industry},
  ISBN = {952-12-1436-8},

Belongs to TUCS Research Unit(s): Data Mining and Knowledge Management Laboratory

Edit publication