|
If
you have read the discussion on the slow stochastic, you might
find the following rather repetitious. The discussion on trading
rules is identical to the slow stochastic, however you may
want to review the computations, since they are the basis
for the slow stochastic values.
Dr.
George C. Lane is the author of the stochastic indicator.
His basic premise is as follows: During periods of price decreases,
daily closes tend to accumulate near the extreme lows of the
day. Periods of price increases tend to show closes accumulating
near the extreme highs of the day. The stochastic study is
an oscillator designed to indicate oversold and overbought
market conditions.
Some
technical analysts prefer the slow stochastic rather than
the normal stochastic. The slow stochastic is simply the normal
stochastic smoothed via a moving average technique.
The
normal stochastic, like the slow stochastic study, generates
two lines. They are %K and %D. The normal stochastic has overbought
and oversold zones. Dr. Lane suggests using 80 as the overbought
zone and 20 as the oversold zone. Others prefer 75 and 25.
Dr.
Lane also contends the most important signal is divergence
between %D and the commodity. He explains divergence as the
process where the slow stochastic %D line makes a series of
lower highs while the commodity makes a series of higher highs.
This signals an overbought market. An oversold market exhibits
a series of lower lows while the %D makes a series of higher
lows.
When
one of the above patterns appear, you should anticipate a
market signal. You initiate a market position when the %K
crosses the %D from the right-hand side. A right-hand crossover
is when the %D bottoms or tops and moves higher or lower and
the %K crosses the %D line. According to Dr. Lane, your most
reliable trades occur with divergence and when the %D is between
10 and 15 for a buy signal and between 85 and 90 for a sell
signal.
Parameters
(14,3)
- Overall
Period - the number of periods used to determine the
highest high and lowest low. FutureSource uses as
default of 14.
- %D
MA Period - the number of periods used to determine
the moving average for the %D value. FutureSource
uses a default of 3.
Computing
the Stochastic
The
first step in computing the stochastic indicator is to determine
the n period high and low. For example, suppose you specified
twenty periods for the stochastic. FutureSource determines
the highest high and lowest low during the last twenty trading
intervals. It determines the trading range for that time period.
The trading range changes on a continuous basis.
The
calculations for the %K are as follows:
%Kt
= ( (Closet - Lown) / (Highn - Lown) ) * 100
%Kt
is the value for the first %K for the current time period.
Closet
is the closing price for the current period.
Lown
is the lowest low during the n periods.
Highn
is the highest high during the n time periods.
n
is the value you specify.
Once
you obtain the %K value, you start computing the %D value
which is an accumulative moving average. Since the %D is a
moving average of a moving average, it requires several trading
intervals before the values are calculated properly. For example,
if you specify a 20 period stochastic, the software system
requires 26 trading intervals before it can calculate valid
%K and %D values. The formula for the %D is:
%DT
= ( (%DT-1 * 2) + %Kt) / 3
%DT
is the value for %D in the current period.
%DT-1
is the value for %D in the previous period.
%Kt
is the value for %K in the current period.
The
values 2 and 3 are constants. You specify the constants and
the length of the time period to examine for the trading range.
Finally,
you may compute the stochastic value using the traditional
smoothing techniques, a normal moving average, or an exponential
moving average. The formulae are not shown here but they are
similar to the formulae for the Exponential Moving Average
(EMA).
|
