| Welcome Guest, please sign in to participate in a discussion. | Search | Active Topics | |
Registered User
Joined: 10/7/2004
Posts: 8
|
I have a rather sloppy version of a 20 period MA pullback. I'm trying to model Linda Raschke's Holy Grail Setup:
"The "Holy Grail" trade was originally described in my Street Smarts book. The setup occurs when the market's trend has been strong enough to cause a 14-period ADX to rise above 30. When the price then retraces back to the 20-period EMA, odds favor a retest of the most recently formed high or low"
My version of the pullback to 20 MA tries to be a bit more liberal: (I've omitted the ADX part of the PCF)
((C1 > AVGC20.1) AND (C < AVGC15 AND C > AVGC25)) OR ((C1 > XAVGC20.1) AND (C < XAVGC15 AND C > XAVGC25)) OR ((C1 > AVGC20.1) AND (C < AVGC15 AND C > AVGC25)) OR ((L1 > AVGL20.1) AND (L < AVGL15 AND L > AVGL25))
Could you please help, maybe tightening up my PCF. Any other suggestions you may have, of a PCF that may work more efficiently than mine, would be very appreciated.
Thank You
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
I can shorten your Personal Criteria Formula somewhat:
(C1 > AVGC20.1 AND ((C < AVGC15 AND C > AVGC25) OR (C < XAVGC15 AND C > XAVGC25))) OR (L1 > AVGL20.1 AND L < AVGL15 AND L > AVGL25)
But I don't know what you want as far as tightening. The trainers can't give setting, interpretation or investment advice. I can help you implement your unambiguous objective definition of a pullback, but I can't give you a definition or do much to help you refine your definition.
How to create a Personal Criteria Forumula (PCF)
PCF Formula Descriptions
Handy PCF example formulas to help you learn the syntax of PCFs!
Understanding Moving Averages - Part I, The Basics
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
Registered User
Joined: 10/7/2004
Posts: 8
|
Thanks Very helpful. I understand you are limited in what adivce you are allowed to offer.
I guess what I'm trying to achieve, is to design a PCF that will signal a pullabck to the 20MA(EMA), when Close is above the MA. The way I have it structured will often times not give a signal beacuse I am relying on both the 15 and 25 MA as boundaries. This works fine when stock is trending higher, but misses when stock is coming off bottoms.
Sorry if I am rambling, but just trying my best to explain.
Anyway once again thanks for your help
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
One suggestion then that would not be out of bounds considering the fact that the 20-Period Moving Average is Exponential would be to make sure oll of the Moving Averages used in the Personal Cirteira Formula are Exponential as well:
(C1 > XAVGC20.1 AND C < XAVGC15 AND C > XAVGC25) OR (L1 > XAVGL20.1 AND L < XAVGL15 AND L > XAVGL25)
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
Registered User
Joined: 10/7/2004
Posts: 8
|
Funny you mention that, I use both simple and exponentional MA's in my PCF to broaden my approach.
Can you just use either the 20 MA or 20 EMA and have a PCF signal when close/low approaches, just above or below the MA.
Anyways once again Thanks
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
Again, I'm limited in what I can suggest, but let's make some assumptions based on your existing formula.
Near MA20 is defined as being within the Range of MA15 through MA25.
Approaching is defined as being closer during the current Bar than the previous Bar without crossing.
Using these definitions, we could create a formula for the Close being near and approaching its 20-Period Simple Moving Average as:
SGN(C - AVGC20) = SGN(C1 - AVGC20.1) AND ABS(C - AVGC20) < ABS(C1 - AVGC20.1) AND ABS(AVGC15 + AVGC25 - 2 * C) <= ABS(AVGC15 - AVGC25)
A similar formula for the Close being near and approaching its 20-Period Exponential Moving Average would be:
SGN(C - XAVGC20) = SGN(C1 - XAVGC20.1) AND ABS(C - XAVGC20) < ABS(C1 - XAVGC20.1) AND ABS(XAVGC15 + XAVGC25 - 2 * C) <= ABS(XAVGC15 - XAVGC25)
A formula for either of the above being True would be:
(SGN(C - AVGC20) = SGN(C1 - AVGC20.1) AND ABS(C - AVGC20) < ABS(C1 - AVGC20.1) AND ABS(AVGC15 + AVGC25 - 2 * C) <= ABS(AVGC15 - AVGC25)) OR (SGN(C - XAVGC20) = SGN(C1 - XAVGC20.1) AND ABS(C - XAVGC20) < ABS(C1 - XAVGC20.1) AND ABS(XAVGC15 + XAVGC25 - 2 * C) <= ABS(XAVGC15 - XAVGC25))
A version of this longer formula for the Low instead of the Close would be:
(SGN(L - AVGL20) = SGN(L1 - AVGL20.1) AND ABS(L - AVGL20) < ABS(L1 - AVGL20.1) AND ABS(AVGL15 + AVGL25 - 2 * L) <= ABS(AVGL15 - AVGL25)) OR (SGN(L - XAVGL20) = SGN(L1 - XAVGL20.1) AND ABS(L - XAVGL20) < ABS(L1 - XAVGL20.1) AND ABS(XAVGL15 + XAVGL25 - 2 * L) <= ABS(XAVGL15 - XAVGL25))
A formula for either the longer version of the Close or the longer version of the Low being True would be:
(SGN(L - AVGL20) = SGN(L1 - AVGL20.1) AND ABS(L - AVGL20) < ABS(L1 - AVGL20.1) AND ABS(AVGL15 + AVGL25 - 2 * L) <= ABS(AVGL15 - AVGL25)) OR (SGN(L - XAVGL20) = SGN(L1 - XAVGL20.1) AND ABS(L - XAVGL20) < ABS(L1 - XAVGL20.1) AND ABS(XAVGL15 + XAVGL25 - 2 * L) <= ABS(XAVGL15 - XAVGL25)) OR (SGN(L - AVGL20) = SGN(L1 - AVGL20.1) AND ABS(L - AVGL20) < ABS(L1 - AVGL20.1) AND ABS(AVGL15 + AVGL25 - 2 * L) <= ABS(AVGL15 - AVGL25)) OR (SGN(L - XAVGL20) = SGN(L1 - XAVGL20.1) AND ABS(L - XAVGL20) < ABS(L1 - XAVGL20.1) AND ABS(XAVGL15 + XAVGL25 - 2 * L) <= ABS(XAVGL15 - XAVGL25))
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
Registered User
Joined: 10/7/2004
Posts: 8
|
Perfect, exactly what I was looking for. Thanks for all your help
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
You're welcome.
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
|
Guest-1 |