Pyramiding into a trend

[dbphoenix]

Quote from ElectricSavant:

I just want to get your interest and others. My motivations are honorable. Sometimes Db the best must be persuaded. Please feel free to contribute now that I have your attention.

Michael B.

Why would I want to do that since you've already started making cracks? It's not worth the hassle.

If you want to know how I pyramid, look it up.

 

[ElectricSavant]

Db you are truly cheating ET and the readers of this thread of your knowledge. I am sorry and we are the losers (you and I)

Michael B.

Quote from dbphoenix:

Why would I want to do that since you've already started making cracks? It's not worth the hassle.

If you want to know how I pyramid, look it up.

 

[ElectricSavant]

Well, I am sorry to hear about your conflict with the administrator of this site, but hey you are still here.

This bold statement you are making shows me the confidence that you have on the varioius ways to pyramid. I think there is a wide spectrum of interpretation that would make backtesting impossible for all scenario's

Michael B.

Quote from MAESTRO:

Hi there, long time since we talked. I had to change my alias due to my personal conflict with baron, oh, well ...

Anyway, if it helps, I could tell you that we have tested Pyramiding versus non-pyramiding in our office using more than 50 diff. trending algorithms. The results were showing us that non-pyramiding is better. It has almost all the statistics better then pyramiding.

Cheers.

 

[MAESTRO]

Quote from ElectricSavant:

Well, I am sorry to hear about your conflict with the administrator of this site, but hey you are still here.

This bold statement you are making shows me the confidence that you have on the varioius ways to pyramid. I think there is a wide spectrum of interpretation that would make backtesting impossible for all scenario's

Michael B.

We have a staff scientist in our company that was assigned to this task. He used quite sophisticated tools and spent about half a year testing diff. pyramiding strategies. We needed it for our own fund. His recommendations were:

1. Do not pyramid
2. Do not average.

 

[ElectricSavant]

I saw your deleted post db...I simply was telling you about my habits. Haven't you and I previously discussed this? I thought you became frustrated and had to delete your post after my edit. At our age it is sometimes difficult to remember at times...

Michael B.

Quote from dbphoenix:

Saves time to do all that before posting in the first place. And why the swipe? Eager to get this thread off to the usual ending?

 

[dbphoenix]

Whatever.

 

[jbt]

Quote from MAESTRO:

We have a staff scientist in our company that was assigned to this task. He used quite sophisticated tools and spent about half a year testing diff. pyramiding strategies. We needed it for our own fund. His recommendations were:

1. Do not pyramid
2. Do not average.

M - did he test parabolic averaging?
1, 2,4,8 etc?

 

[MAESTRO]

Quote from jbt:

M - did he test parabolic averaging?
1, 2,4,8 etc?

I'm not sure, let me ask him.

 

[Maverick1]

Quote from MAESTRO:

I'm not sure, let me ask him.

Success with pyramiding is conditional on the entry system

Saying that pyramiding doesn't work without defining the entry system against which the pyramiding was tested doesn't seem to be a sound proposition, logically speaking

 

[ElectricSavant]

Without revealing what was shared with me in a PM, as I have no right to claim these are my thoughts.

BUT....

trend must be identified in stages...I knew this. But it has been shared with me that it is a tradable valid methodology to use different methods for different stages (the chameleon comment).

Now lets not talk too much about Elliot Waves...but I see a connection.

Michael B.