PureBytes Links
Trading Reference Links
|
Thanks for the emails.
RS systems has an example that is easy to understand. i.e., the process of
going from raw data to processed data.
http://www.rs-systems.com/example.html
===========
from the FAQ's
" ... What is knowledge discovery?
Knowledge discovery is the search for hidden regularities in databases.
These hidden regularities are a basis for correct decisions. Knowledge
Discovery tools find these trends and generate rules to explain the
regularities. Future decisions can be based on the rules generated from
databases.
What are Rough Sets?
Rough Set Theory, introduced by Zdzislaw Pawlak in 1981, is a new
mathematical tool managing uncertainty and vagueness in data sets. The main
concept of rough set theory is an indiscernibility relation, normally
associated with a set of attributes. Sets that are indiscernible are called
elementary sets. Any finite union of elementary sets is called a definable
set. Two notions are fundamental: lower and upper approximations of a set.
The lower approximation of a set is the set being the union of all
elementary sets contained in the given set. The upper larger approximation
of a set is the smallest set being the union of all elementary sets
containing the given set. ...etc."
==========
reference to Ruggiero's and Skalkos' work is at
http://www.reduct.com/Applications/Business/business_examples.htm#OEX
Chris Skalkos, "Rough Sets Help Time the OEX", NeuroVe$t Journal.
November/December 1996, pp. 21-28
NEUROVE$T JOURNAL
TECHNICAL ANALYTICA
TECHNICAL MARKET ANALYSIS & INSIGHT
ROUGH SETS HELP TIME THE OEX
CHRIS SKALKOS
This paper describes an application of rough sets analysis to trading the
OEX. An essential goal of this analysis is to determine when a trader should
fade an indicator, therefore going opposite to prevailing market sentiment,
or join its trend. Using rough sets techniques, a set of rules for
short-term trading the OEX based on the Hines indicator is extracted. A
System is then developed, encompassing all of the derived rules, in order to
evaluate trading System
"... 2. DATA PREPARATION
We start by examining the behavior of the Hines indicator over the time
period from October 8, 1987 to August 7, 1995. Over this interval of time
the indicator has a mean of 1.1014, implying a slightly bullish long-term
sentiment bias, with a Standard deviation of 0.3084. A reading of 0.793,
which is one Standard deviation below the mean,
implies considerable bearishness. Such readings occurred 242 times during
the aforementioned period, out of which 157 times it correctiy signaled an
imminent upward move in the market, representing a success rate of 64.8%. A
reading of 1.4098, which is one Standard deviation above the mean, implies
considerable bullishness. Such readings occurred 272 times, out of which 144
times it correctly signaled an imminent downward move, representing a
success rate of 52.9%.
Figure l presents the OEX prices and Hines indicator readings over period
July 1993 through March 1994. The Hines indicator is shown below the OEX
time series along with three horizontal lines indicating plus-one sigma,
mean and minus-one sigma, respectively. Readings of less than 0.793 during
mid-July, November and Decem-ber of 1993 were followed by a higher market.
Figure 2 presents the data for the period covering mid-June 1991 through
February 1992. Readings greater than 1.4098 during August, the end of
September and the beginning of November were followed by a lower market.
There are periods, of course, where repeated extreme readings of the Hines
indicator are not necessarily followed by a short-term reversal. Figure 3
shows such a period, covering November 1994 through July 1995, where low
readings during March, April, May and June of 1995 indicated excessive put
buying as OEX options speculators fought the upward trend of the market.
Our rough sets analysis will attempt to identify under what conditions
extreme readings of the Hines indicator prognosticate either an imminent
reversal or a continua-tion of the market's momentum.
...etc"
Ruggiero's work appears in Futures Magazine regularly.
Best regards
Walter
|