Signal clarity

[ACM]

David
Attached is the framework efs with the MAs of indicators each one with its own function parameters.
One note regarding the MAofStoch. Since there is a bug which does not allow to directly calculate the MA of StochStudy.SLOW I have implemented a workaround that essentially computes the MA of the MA of StochStudy.FAST.
At this point could you please transfer all the logic of the latest FGallFalse.efs to this one? I would still leave in both strategies ie the original OB/OS and the modified "contrarian" one.
Lastly, the best results I have had so far across various periods seem to be using the contrarian with 28/14/16-1-3 respectively for CCI, RSI and Stoch.
Alex

[ACM]

David
I believe there is an error in Line 74 of the efs I posted earlier. You may want to check it but I think the line should read
if (vMAofStoch == null) vMAofStoch = new MAStudy(Slow, 0, vStoch, StochStudy.FAST, MAStudy.SIMPLE);
Alex

[dloomis]

Re: Reply to post 'Signal clarity'

FG

I've been away for a few days - any thing new?

How do the changes Alex suggested look?

DO you think MA's of the cci etc. will be something to look at?

Have you tried any thing new?

Any program changes we should make?

Let's not get to nervous about the results of any on specific trade - i.e.
short from 1072 - let's focus on the long term and getting some signals that
turn with the market


----- Original Message -----
From: <[email protected]>
To: <[email protected]>
Sent: Thursday, December 18, 2003 10:00 PM
Subject: Reply to post 'Signal clarity'


> Hello dloomis,
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>

[FibbGann]

I just got back myself and was looking at what Alex did. I think if its not to much trouble we should go with the MA set-up Alex has prepared. As to this way we are not having to jump in and out all the time. I visually seen better signals using the MA's. I will have time over the holidays to run multiple tests because I already had mine early.

Let me know.

[FibbGann]

I have a few other ideas I will try. If you could transfer the logic into Alex's updated frame work I will run a series of strategies and test. I will have all the testing done by the new year. I don't plan on trading much til then. Plus, I have a few other projects I am working on. Also if you have anything else you might want me to cover, just let me know.

[dloomis]

Re: Reply to post 'Signal clarity'

I would be glad to add some logic into Alex' frame work.

What did you have in mind? - I seem to be lost at this moment, sorry.

[FibbGann]

Here is the framework Alex did for using MA's.

Refer to Alex's last two posts

[FibbGann]

This is the efs logic I beleive We were using last. Alex correct me if I am wrong.

[FibbGann]

I will take the old theory and Alex's new one (OB/OS) And run tests on both using the MA's.

[dloomis]

Re: Reply to post 'Signal clarity'

FG

Yes, that's the most recent file.

[dloomis]

Syntax errors fixed

[mattgundersen]

Hey All,

I took the FGrsi, FGsto, and FGcci from the ThreeSignals file share group and threw them on the bottom of this chart. This also has fgallfalse(2b).efs and fgallfalse3.efs.

Before I go furhter - I'm kinda confused. What is the proper version of this script to run?

Also, on the FGrsi, FGsto, and FGcci, those are not MA's of those studies, just the raw study. Does anyone have EFS's that take each study from fgallfalse3.efs and breaks them out into individual EFS's/panes? I saw a post many many posts ago that showed them broken out into separate panes and showing the individual signals. Just wondering if anything like that still exists?

One additional thought to this method is adding some kind of price filter.

In the image below, I have a bunch off the EFS's from this thread and am showing some Envelope's on the chart as well as a Donchian. I've been doing some bell curve fitting stuff and measuring which Envelopes tend to be hit most often. On a 3 minute e-mini, I'm seeing that the 0.2618, 0.1618, and 0 (basis) envelopes are most often hit.

Perhaps using an envelope with some %, the basis of an envelope, or the donchian could be used to help qualify or raise the odds on a successful trade. eg: if FG gives a signal and the .1618 or .2618 envelope is hit, chances are good for a reversal.

Thoughts?

And ps: where can I find all the latest FG.efs?

[FibbGann]

I have the MA of CCI and RSi efs's , I do not have an Ma of Stochastics though. There is becoming a little spread of misunderstanding in here of what exactly is going on. (not much though)

P.S Matt Just curious if you had an answer to my e-mail I sent you about ESP?

Fibbgann

[FibbGann]

Here is the other efs Matt. Alex? when you coded the Fgallfalse (2b).efs. Were you intending to use the median of each indicator for the B/S signals. AKA

RSI over 50 sell
RSI under 50 buyl

CCI over 0 sell
CCI under 0 buy

STO over 50 sell
STO under 50 buy ???

This is the only thing I am not sure of at this point.

Fibbgann

[FibbGann]

Picture