| Welcome Guest, please sign in to participate in a discussion. | Search | Active Topics | |
Registered User
Joined: 10/7/2004
Posts: 1
|
i am looking how to write pcf for a 10 day sideways base
thank you sanbro
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
Welcome to the forums. A very good foundation for learning how to use TeleChart can be gained by reviewing the following:
If you are new to TeleChart READ THIS FIRST!
Do you have a particular unambiguous objective definition of "sideways base" in mind?
One option might be to compare determine the Average Range over 10 Days as a Percentage of the Total Range over the same Period. The higher the Value (the closer it is to 100%), the more "sideways" the movement (-100 could also represent sideways as well if it is a flatline - I did that on purpose to differentiate it from other "sideways" movement):
100 * ((AVGH10 - AVGL10) / (MAXH10 - MINL10 - (MAXH10 = MINL10)) + (MAXH10 = MINL10))
Another option might to determine the Annualized Price Growth Rate over those 10-Days based on the Linear Regression Slope. The closer it is to zero, the more "sideways" the movement.
100 * (EXP(253 * (4.5 * LOG(C ) + 3.5 * LOG(C1) + 2.5 * LOG(C2) + 1.5 * LOG(C3) + .5 * LOG(C4) - .5 * LOG(C5) - 1.5 * LOG(C6) - 2.5 * LOG(C7) - 3.5 * LOG(C8) - 4.5 * LOG(C9)) / 82.5) - 1)
You may wish to review the following:
Scan for stocks bouncing in a lateral channel
How to create a Personal Criteria Forumula (PCF)
Using Linear Regression vs Classical Peaks/Valleys for Divergence Analysis
PCF Formula Descriptions
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
Registered User
Joined: 7/23/2010
Posts: 84
|
Bruce how could we make a PCF for a 5-6 week (Weekly Flat Base) whereas the bottom of the base is flat (the top of that base is not so much of an issue as is the bottom should be quite flat. I'd like to use TC 2000 if possible. But I suspect you might say you are not allowed to provide support for a beta version. If so then in SF5 (or even TC2007 & I might be able to take the spaces out for a conversion to TC2000).
Thanks for your help & advice
art
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
I can help you create something for our current releases (not future releases).
You need to come up with an unambiguous objective definition for "flat base" to be able to identify it however. You might know that the bottom is "quite flat" when you see it, but we need to be able come up with some algorithmically identifiable fact that is true when this is case but not true when this is not the case.
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
Registered User
Joined: 7/23/2010
Posts: 84
|
From another link which you provided awhile back I input this
Trend/Consolidate 3-month
100*(MAXH63-MINL63)/(AVGH63-AVGL63)
Which comes close to my 5 month goal. How could this be modified for 5 months? Also I would like to find stocks (which show true) which had been in an uptrend of 2months or more and have then entered into the above (& modified to 5mo) consolidation period.
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
5-Week or 5-Month (you have used both)?
Trend/Consolidate 5-Week:
100 * (MAXH25 - MINL25) / (AVGH25 - AVGL25)
Trend/Consolidate 5-Month:
100 * (MAXH105 - MINL105) / (AVGH105 - AVGL105)
Just as with a base, we would need an unambiguous objective definition of uptrend to be able to identify it.
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
Registered User
Joined: 7/23/2010
Posts: 84
|
Bruce PCLN comes to mind (see weekly chart) which has been in an uptrend from June and last 4-5 weeks consolidating.
|
|
Worden Trainer
Joined: 10/7/2004
Posts: 65,138
|
While examples can be useful in clarifying an already objective definition, they are on their face subjective (you know it when you see it). We need an unambiguous objective definition of what you are attempting to identify to be able to actually algorithmically identify it using a program.
-Bruce
Personal Criteria Formulas
TC2000 Support Articles
|
|
|
Guest-1 |