Ichimoku5

newdigital · #1 (**permalink**) 10-31-2005, 01:35 PM

I like this EA. It is great. Thanks Foreverold

You may find this EA in down load area. It is for MT4.

But I think it is necessary to improve something.
Just a something.
See description here here

forexts · #2 (**permalink**) 10-31-2005, 01:53 PM

If you backtest only last month ... you will see ea lost a lot of money this month (In average of other months profits are very good).

Beluck · #3 (**permalink**) 10-31-2005, 07:20 PM

which model did you backtest it?

if not "open prices" and do not have low level timeframes in your metatarder for the whole period of time then you may get unrealistic results.

i think for the last month you see more or less real test as you have data for small timeframes, which are used for modelling bars of larger timeframes.

read articles here with explanations:
http://www.metaquotes.net/experts/articles

newdigital · #4 (**permalink**) 10-31-2005, 07:32 PM

I backtested on h4 every tick. Of course I did not have the m5 and m15 etc for the whole period of h4.
I tried to doanload the data from alpari but there are a lot of gaps there.
Backtesting results is here Ichimoku EA

But I am continuing testing it and seems this EA needs to be improved.
See my statement above.

newdigital · #5 (**permalink**) 11-15-2005, 11:37 AM

I found the reason why I got very strange results with this EA on h1 timeframe (two orders was opened and closed within 1 minutes). And it was not backtesting. It was live testing on demo (forward test).

Thanks Codersguru for the lessons.
I found in the code of this EA the following:

Code:

double diCustom0=iCustom(NULL, 240, "Ichimoku", 9, 26, 52, 0, 0);
double diCustom1=iCustom(NULL, 240, "Ichimoku", 9, 26, 52, 1, 0);
double diCustom2=iCustom(NULL, 240, "Ichimoku", 9, 26, 52, 1, 0);
double diMA3=iMA(NULL,240,34,0,MODE_EMA,PRICE_CLOSE,0);
double diCustom4=iCustom(NULL, 240, "Ichimoku", 9, 26, 52, 1, 0);
double diCustom5=iCustom(NULL, 240, "Ichimoku", 9, 26, 52, 0, 0);

So, 240 is the timeframe. Foreverold was right here.
This EA was design for the H4 timeframe.

Thus I changed something in the code and we may select timeframe, working hours and Ichimoku setting (Tenkan_sen, Kijun_sen and Senkou_Span_B).
Everything about Ichimoku settings you may find here

I uploaded this Ichimoku 5_1 EA to the download area. Codersguru may check my coding for any mistakes and everybody can play with timeframe, Ichimoku settings and working hours to find the right settings for any timeframe except H4. For H4 this EA is very good and nothing can be changed for h4 timeframe. May be H1 is ok as well.

tokyowalker · #6 (**permalink**) 12-26-2005, 07:43 AM

hi everybody ..i am new here and i really like this forum ..thanks for keeping the good work ..
i started testing the ichimoku5 4 hours time frame for (EURUSD, EURAD, GBPJPY, GBPUSD) in a demo account ..the results were good for the first day but no luck for todays results

this EA is trading the hole time and i think it needs some improvement ....

newdigital · #7 (**permalink**) 12-26-2005, 11:12 AM

Quote:

Originally Posted by tokyowalker

hi everybody ..i am new here and i really like this forum ..thanks for keeping the good work ..
i started testing the ichimoku5 4 hours time frame for (EURUSD, EURAD, GBPJPY, GBPUSD) in a demo account ..the results were good for the first day but no luck for todays results

this EA is trading the hole time and i think it needs some improvement ....

You tested it during the 2 and half days only. It is nothing for this EA.

This is always-in-play EA. It does not mean that it is very profitable. The profit may be the same with any other EAs (in average).

For example you will have for one week +150 in profit and -120 losses. +30 in average.

But there are EAs which are not trading often: just the same +30 with no losses.

So diference is that this EA is always-in-play.

Some people like always-in-play EAs. Why? Because this EA is always trading and you will not miss your profit (order was already opened). Other EAs are trading one time per week during the short period of time (for example) but with the same profit. What is more risky: be always together with the price playing or trade just sometimes (with the same profit for example)?
I think it is up to the people.

tokyowalker · #8 (**permalink**) 12-26-2005, 04:12 PM

thank you newdigital for your reply ..i will keep the test doing its job for one week and post the result again :d as i told you i m new here and i just started getting familiar with what s going on in the forex world ...
thanks again

geneometrik · #9 (**permalink**) 12-27-2005, 08:58 PM

Been using Ichi5 on areal ,mini real for a week...found a bug. When I put lot size (.10) it actually buys a $1,000 contract. or 10 cents per pip. When I leave it @ default 1 Lot it says there is not enough money in the account. I'm assuming it's trying to buy 100K lot. So the question is How do you buy a standard mini lot which is actually $10K?

Need help on this one fellas. Also been using on 5 min with wider trails and longer limit runs...It prervents the EA from over trading... I really dig this one...one of my favorites. And please if any one could please help with the mini lot situation, I would be greatful.

Thanks in advance,
G

tokyowalker · #10 (**permalink**) 12-30-2005, 06:18 AM

hi guys
here is my ichimoku tesing statment on demo from (2005.12.27 01:17 ---- to----2005.12.30 00:00)...used time frame is of course 4H and used currency pairs are (GBPJPY-GBPUSD-EURAUD-EURUSD)..it was running 24hours a day without any human interference...i chosed this pairs cause they had unbelivable results on backtesting ..but it looks like you should never trust backtesting

anyway the bottom line for each pair (1lot) :
GBP JPY = -1877,47 a huge loss

EUR AUD = -491,83 another huge loss

EUR USD =-1080 my god that s another freaking huge loss

GBP USD = + 450 (yeah but ...)

anyway i think i ll continue testing it but just on the GBPUSD for a another week ...just to make sure that it might be a winning pair ..

one remark : maybe it s not a good idea to let the EA open a trade on the same pair right after a big loss on that pair , cause as you can see on the statment there was some successive big losses onthe same pair and the system just doesn t care and open another trade with the same direction... this EA is doing great on backtesting by he way ...

the statment is in the attached file ...