Dear IIF, TS and OR discussion groups,
In OR milestones collection at
http://www.lionhrtpub.com/orms/orms-10-02/historysb1.html the
pioneering book by Zadeh titled "Fuzzy Set Theory" (FST) appears
immediately after "Smoothing, Forecasting and Prediction of Discrete
Time Series" by Brown. On the other hand, forecasting using fuzzy set
theory emerges without glory from B. S. EVERITT in the Second Edition
of THE CAMBRIDGE DICTIONARY OF STATISTICS. See part of the entry
below. On the same hand, modern reviews (e.g., "25 years of time
series forecasting" by Jan G. De Gooijer and Rob J. Hyndman) seem to
mention no articles in the leading forecasting journals that would
connect FST to TSF. (Fiordaliso, A. (1998) tends to be more relevant
to combining forecasts) I currently have only two more or less
relevant papers:
1) A Fuzzy Forecasting Model for Apparel Sales
http://www.ntcresearch.org/pdf-rpts/Bref0602/S01-PH10-02.pdf
Forecasting Women’s Apparel Sales Using Mathematical Modeling
(this one involves Winters’ Three Parameter Exponential Smoothing
Model)
http://faculty.philau.edu/frankc/ntc/index_files/Report1CF%206%20301.doc
Finally published in "International Journal of Clothing Science and
Technology"
2) FUZZY BASED TIME SERIES FORECASTING OF ELECTRIC LOAD P. Otto, T. Schunk
http://www.zstu.zaporizhzhe.ua/base/ric/files/riu21999/99_2_3.pdf
(appeared in a Ukrainian journal, in 1999, though in English; Abstract
starts: "The paper presents a new method for the forecast of time
series.")
An article to buy:
http://www.actapress.com/PaperInfo.aspx?PaperID=21093
I wonder if there is any solid documentary evidence or expert's
opinion to back up or refute the last statement of the entry below.
Any commentary, relevant links, references, doubts etc welcome. FST
for TSF is of general interest to me. I was taught to take for granted
no single source of scientific information. Hence this call for a
discussion. Thanks.
===
Fuzzy set theory: A radically different approach to dealing with
uncertainty than the traditional probabilistic and statistical methods.
The essential feature of a fuzzy set is a membership function that
assigns a grade of membership between 0 and 1 to each member of a set
A. Mathematically a membership function of a fuzzy set A is a mapping
from a space X to the unit interval m_A : X --> [0,1].
...
Methods derived from the theory have been proposed as alternatives to
traditional statistical methods in areas such as quality control,
linear regression and FORECASTING, although they have NOT met with
universal acceptance and a number of statisticians have commented that
they have found NO solution using such an approach that could not have
been achieved as least as effectively using probability and statistics.
===
---
Best wishes,
Andrey Kostenko
---
Quote of the day:
The fact that an opinion has been widely held is no evidence whatever
that it is not utterly absurd.
Bertrand Russell
|