INDICATORS
A Wilder Variation
Reverse Engineering RSI
by Giorgos Siligardos, Ph.D.
Can the inverse of an oscillator help you to forecast closing prices?
Reverse engineering is a mathematical procedure
that takes the inverse of an oscillator. In this article I will illustrate
this process using J. Welles Wilder's relative strength index (RSI). The
reverse-engineered RSI, or RevEngRSI for short, can help determine the
following time period's closing price using the value of the oscillator.
RSI REVISITED
To define the RSI of k periods for a daily graph, it is necessary to
first define the up-close and down-close indicators (UC and DC) for the
day n. These are as follows:
where Cn and Cn-1 are the closing prices for the days n and n?1.
In the sequence, the average up close of k periods [Auc(k)] is defined
to be the (2k-1)-period exponential moving average of UC. The average down
close of k periods [Adc(k)] is the (2k?1)-period exponential moving average
of DC. The RSI of k periods for day n then is:
The RSI of k periods may be defined for weekly, monthly, or annual graphs
by substituting the daily closing prices with weekly, monthly, and annual
closing prices. Wilder favored a 14-period RSI. In her book Technical
Analysis For The Trading Professional, Constance Brown states that
she also prefers a 14-period RSI because of specific forecasting capabilities.
However, while Brown spends an entire chapter in her book on reverse engineering,
she does not give the formula for reverse-engineering the RSI.
...Continued in the June 2003 issue of Technical Analysis
of STOCKS & COMMODITIES
Excerpted from an article originally published in the June 2003 issue
of Technical Analysis of STOCKS & COMMODITIES magazine. All rights
reserved. © Copyright 2003, Technical Analysis, Inc.
