|
PureBytes Links
Trading Reference Links
|
Terry,
I've been brushing up on statistcis recently. Why do you use 3 standard
deviations? Wouldn't 2 be sufficient?
-----Original Message-----
From: TQuinn211@xxxxxxx <TQuinn211@xxxxxxx>
To: RealTraders Discussion Group <realtraders@xxxxxxxxxxxxxx>
Date: Monday, July 13, 1998 9:33 PM
Subject: Re: leading/lagging indicators
>
>The ela file attached contains 4 indicators. Two from Walt Downs and two
from
>me.
>
>Walt's provide the full moon dates and counts the number of turns within a
>lunar year. If you have any questions regarding these two indicators Please
>contact Walt.
>
>The indicator marked as X-bar is my first indicator.
>
><<
> The first is a moving average channel based on the work of Dr. Edward
Deming
> who never traded a stock/index or Future in his life, but was recognized
as a
> foremost statistician in his lifetime. (as a side note I am not surprised
> Japan is in an economic mess since his death.)
> >>
>
>This indicator gives three plots.
>
>Daily price average of open/hi/low/close. (xbar)
>
>A twenty-five day period average of the daily price average. (x doublebar)
>
>Moving average envelopes for the Xbar based upon the average daily price
range
>of the last 25 bars. (mutiplied by a specific factor that is statistically
>derived based on sampling theory. As the sample size changes so does the
>multiplying factor.).
>
>These envelopes provide the most efficient +/- 3standard deviations for the
>Daily price action as represented by the daily xbar.(Sorry John Bollinger,
you
>almost got it right.)
>
>It is specifically designed to answer The Most Important Question needed to
>design a sucessful trading system. What is the trend? Up, Down or
sideways?
>
>Here is a rule. If three X-bar's are outside the limits in a particular
>direction. a trend is in the making.
>
>Test it and I think most of you will find it to be a most superior moving
>average trading system.
>There are also rules to tell one when a trend is over. Those will be in my
>book if i ever write one.
>Since it takes three plots to identify the trend, It is a Lagging
indicator.
>
>
>The other indicator is called linslope2. The basic principle behind this
>indicator is based upon a simple concept in physics. When is a system in a
>steady state and when is it changing?
>
>
>Up, down or sideways if a system is moving at a constant rate of speed then
>the rate of change to the system is zero. Think of a car. It can be
standing
>still or moving at a steady rate of 60 miles per hour. the rate of change
>would be the same. "zero" But if we accellerate or decellerate the car then
>the "Rate of change for speed" would oscillate.
>
>That is what this indicator measures.
>
>The two lowest plots on the chart are for the same indicator. Only the
Number
>of days have been changed. Note how the 10 day(red) turns faster than the
>yellow (13 days).
>
>The 10 day also turns before price, giving it an opportunity to be a
leading
>indicator.
>
>
>I have found that this indicator works well as a divergence indicator.
(Higher
>high in price and lower high in the indicator) and as a screening mechanism
>for other oscillator entrance rules.
>
>As an example if a stochastic turns up and this indicator is turning up
from
>an "oversold" location then I give the Stochastic signal more weight.
>
>On the otherhand if there is a stochastic buy signal and the LINSLOPE2 has
>already passed it's zero line, I will most likely pass on the trade.
>
>Hint:You folks doing cycle work may want to try !/2 the average cycle
length
>for this indicator an then look at your buy signals in relation to this
>indicator.
>
>
|