|
Jim Dean
 Sage
  Posts: 3022
Joined: 9/21/2006
Location: L'ville, GA
User Profile |
Just watched Jeff’s seminar. Very well done. This appears to be an excellent step forward.
My primary question is about how to “balance” the selections from the various enabled market state groups. I suspect that the Sequential method is the best vehicle for that.
Jeff clearly explained how it rotates thru each Group and works from the top down, peeling off symbols from each group until it runs out of money.
Here is a scenario based on that, presuming three groups each with four states as per his slide. Let’s assume that each state has its own internal filtering - maybe associated with the rankings, maybe something else - so that the following number of symbols pass the test on a given bar:
Group 1: 3 symbols avail a b c
Group 2: 1 symbol avail d
Group 3: 9 symbols avail e f g h i j k l m
The process will select, presuming funds are available, a d e b … then (presumably) f c g. After that, if funds are still avail, I presume it selects h i j k l m.
Ok - if that’s not correct then please tell me how it does work for that example.
That’s the scenario. Now the QUESTIONS:
1. How can I get it to automatically stop when it hits a pass that there are not symbols available in all three groups (or maybe in at least two of the three?
2. How can I get it to keep track of the Capital &/or Risk in each of the positions, so that it keeps on going with more symbols so optimize a rough balance of funds committed to each group (or to risk taken)?
Obviously 2 is tougher than 1 - but it would be wonderful if possible.
|
|
FG
 Member
 Posts: 39
Joined: 1/8/2004
Location: Atlanta, Ga
User Profile |
Jim, It seems to me the key is how the ranking system works. Does it generate ranking values by comparing all trade candidates in all groups collectively? On the other hand, if it ranks one group at a time by comparing trade candidates only in that group, trades entered sequentially could vary widely in rank missing higher rankings because funds were exhausted before re-sequencing.
[Edited by FG on 6/29/2018 11:15 AM]
|
|
Jim Dean
 Sage
  Posts: 3022
Joined: 9/21/2006
Location: L'ville, GA
User Profile |
Yes. That too.
I was presuming normalized ranking with high pass filters in each Strat.
[Edited by Jim Dean on 6/29/2018 12:32 PM]
|