Se-fu Ichimoku Envelope Trend revised

Post by **Jonathan2FI** » Sat Feb 15, 2014 4:37 pm

As a further development I have been looking at the types of trends that can be identified with Ichimoku
1. The first EA addresses established trends bouncing off the cloud
2. The other type you see is stronger momentum moves where the Chinkou line moves strongly in one direction and price does not retrace back to the cloud. See pic

I have developed an ea for this type that appears to be stronger than the other EA I will put onto post 1 after further testing. An interesting aspect of this latter EA is the use of Target Profit defined as the the pips between Chinkou (momentum) line and close all at 26 bars back to set Target Profit. It is obvious but I will state it if you develop a system that has no TP and you simply run it to stop loss or the reversal of entry conditions then you will be closed out at an extreme- hopefully after it has moved enough pips to sustain this loss of pips. Since hope is not a strategy I have looked at TP in detail.
Note Pivots are a great way to set TP's but for trend trading they have their limitations.
Jonathan

AGT · Post by **AGT** » Fri Feb 21, 2014 8:42 pm

Hi Jonathan - any good news here from your side ? How does this ichi thingy work now ? Any MFXB links to see the results at a glance ?

Post by **Jonathan2FI** » Sun Mar 02, 2014 7:03 pm

AGT,
Thanks AGT I am at a final iteration (ran out of weekend time so on hold until next week) and will provide final update plus 7 years updates plus set up forward testing on demos.
Thanks for providing your link to testing - shows opportunities and challenges of long term profitability

Any comments from other Ichi or Heikan Ashi traders?
Jonathan

Post by **Jonathan2FI** » Sat Mar 22, 2014 6:34 am

Have updated this EA with some fundamental changes - I have gone back to a version that is robust across different currencies with good long term ratio profit to DD on a single TF.
It also uses the 240LWMA on H4 aka Bob.
I will put on fwd testing on axitrader see updated post 1
Jonathan

AGT · Post by **AGT** » Sat Mar 22, 2014 8:41 am

Nice to hear from you again ! Watching your project with high interest !

Post by **SpiderX** » Sun Mar 23, 2014 4:14 pm

Hi Jonathan,

Is your EA still version 509 ?

Cheers

Post by **zentauro67** » Mon Mar 24, 2014 5:37 am

Take a look.

http://www.babypips.com/school/elementa ... %20Twitter

Post by **Jonathan2FI** » Sat Mar 29, 2014 2:42 pm

SpiderX » Sun Mar 23, 2014 4:14 pm wrote:Hi Jonathan,

Is your EA still version 509 ?

Cheers

Updated 7 days ago and not since
Jonathan

Post by **Jonathan2FI** » Sat Mar 29, 2014 3:27 pm

Hi All
the version on post 1 is latest and appears to have legs over 4 -6 pairs I have tested on.
I looked at Bob and Shelly's great EA development and developed an additional sefuichi version with an additional 60lwma plus a test of 240 60 converging or diverging......... but when I BT it it was not as good. That has nothing to do with the quality of Bob and Shellys EA!. good + very good does not necessarily = excellent ...obvious really.
And Ah but is BT to be relied on? If it is done in a consistent way with good data, std spreads over a long period 3- 7 years for 4h EAs it can produce some statistical evidence...nothing more.
So I would use BT (long term) to filter the development EAs possible from no way.
Apologies for not posting results yet I had a miss start last week that I did not pick up until end of week with a balance filter preventing trading.

Annoying since the EA on post 1 on EURAUD could have netted 10% over the last 2 weeks mmm I did use the word "could" and I reiterate that this is not a fire and forget EA they do not exist.
I will look at development opportunities as well but if you look at many of the posts on this forum by the EA developers coders you will often see a discussion about not developing further or adding complexity - more is not better!

EA cycle 1. Idea (based on sound trading logic - Coding - debugging - refinement - over refinement and death plus a deal of frustration over many hours wasted.

EA cycle 2. Idea (based on sound trading logic - Coding - testing/debugging - testing/refinement - Go live - EA no longer provides edge based on statistical evidence of performance -kill and start development again.

As Mr Long pointed out to us a while a go over development and complexity is a dead end road for an EA.

So lets get some real testing under way. Any help on this would be appreciated
Jonathan

AGT · Post by **AGT** » Sat Mar 29, 2014 5:56 pm

Hi Jonathan ! Nice to see your consistent efforts !

Which pairs are these 4-6 you mentioned ? Any specific setfiles for them exist ? On post #1 only a eurusd setting is listed.
AND: Did you run the bot also during the first 1-2 weeks of march ? Bcz since the middle of march we have nice trending markets - thats ideal for ichimoku breakouts ...