person could probably pick out areas where the price was going
nowhere over a period of time. This perceptual phenomenon occurs
whether we’re looking at tick charts, daily, weekly, or month charts.
And, just for starters, I’m going to circle the areas I pick out.
However, to eventually get this idea of prices being relatively stable
into a computer, I’ll need to be more objective. Clearly lower
lows/lower highs or higher highs/higher lows will not constitute a
trading range. Prices must return to the approximate levels they
achieved in some period before.~ Otherwise, the central value, if
there is, one, is still moving upward or downward.
This return to a previous price level can be observed visually and
by the computer. All the computer needs to know is how far back
to look when making its comparison. As shown in Figure, if we
set that look-back period too short, it won’t see the last low. If we
set it too long, it might pick up a low six months ago.
How far back do we look? If we can see trading ranges on daily,
weekly, monthly, or even five-minute charts, which one is relevant?
This is where the time horizon discussed earlier comes into play.
Time Horizon
First, you’ve already chosen how often you’ll monitor the market,
usually looking at the data for entry signals. A retail speculator has
the luxury of choosing his time frame: intraday, daily, weekly, biweekly
and so on whereas, institutionally, the exigencies of serving
customers or the business often dictate the trading horizon. Although
there exist retail day-traders, most don’t have the daytime
Daily prices give the most data points so I’ll use those.*
Second, you look for cyclic content. Chapter 1 discussed used
maximum entropy spectral analysis (MESA). Another alternative
is Fourier analysis.
Most traders also constrained by how deep their pockets are.
The longer the time frame, the greater the excursion of prices and
the greater the potential losses they face. Traders who can take the
losses from trading daily bars may not be able to handle the swings
on a weekly or monthly basis. Some will luxuriate in having the
depth to handle a longer time frame while choosing to trade in a
shorter time frame, but most traders will intuitively watch the market
as much as necessary to control losses in relation to the size of
their accounts.
Ranging
“Trend” is a word that could use some analytical rigor, especially in
the trading literature. Frankly, there is a lot less to trading than
meets the eye,s and the old Dow rule of higher highs, higher lows
(or the reverse) is what many trading indicators come down to. At
the least, that sort of activity will trigger nearly all indicators looking
for trends.
To trade ranges, we’re looking for the absence of trend within
the time frames of 12 and 58 days. Possible rules, among hundreds,
include
1. n-day range diminishes and volatility declines
2. n-day absolute average slope diminishes and volatility declines
3. not in trend mode, however trend mode is defined
4. absence of higher highs/higher lows or lower lows/lower highs
5. simple crossing of price over trailing average
6. pivot point highs and lows not exceeded for 11 days
You can see that this speculation could go on forever. Having
picked a definition, you’ll test it to see if (1) it occurs and (2) it occurs
enough for trading. If so, we proceed to trading.
Read More : RECOGNIZING RANGING
