Indicators development for T101's not very simple method

----------------------------------------------------------------------
BTW, if anyone remembers, the indicators I made and gave away long ago, I did T101 without offline charts, so the code is there to be picked up and put into EA
DON
----------------------------------------------------------------------

Hello Don -

Do you have a Indicator that combines the 14 currency pairs without being an "Offline Chart"? Backtesting is very difficult and I think it is very UNRELIABLE backtesting on the current "t101_indy_pricehistory_v3-dnk_donfix14_hopfi2k_zero_div_fix tick_fix"

Steve

Thanks a lot Don!! You definately got my attention

I doubt I'll be one of your "scientific testers" - the conditions of my setup are just too unstable in the current condition. Having said that I'll definately get somthing set up and running and will let you know how it goes.

Thanks again Don!

Regards
Gary

Hey Don,
You certainly deserve a well earned rest! That is a massive effort.
Was just trying to understand the templates, and have started with the m1m15_20LWMA template, but not sure which EA should be dropped on it.
Or, can all the templates accept either EA?
Thanks,
EP

Its OK, I got it. Had to change one indicator to "true" in the inputs.

Thank you Don. I want to be a tester for VQzz and VQ7.

I am running 4 EAs now.

VQzz M1
VQzz M15
VQ7 M1
VQ7 M15

All 4 on EURUSD.

Can somebody attach the indicator:

t101_indy_pricehistory_v3-dnk_donfix14_hopfi2k_zero_div_fix tick_fix

Thanks,

Y.

Here you are

Just a quick note. On setting up Don's EA's I noticed that you need to put the indicator which applies to indicator you select in the EA inputs into the indi folder along with the one attached here: I saw it being called for in the experts tab - you need it.

GG

Steve,

Here's the indicators I've uploaded to this thread that's relevant to the discussion.

First came my BuyAll indicator, modified by my friend Stella and uploaded to another forum, where Orest found it and posted it here, which brought me to posting on the thread. To get the BuyAll, which idea offline chart came from, use BuysVsSells and choose to show on BuyAll line, or I guess later IA I gave would be best? I don't know. I know I've fixed the first bar of the week, thank God, and no longer want GMT offset, just whatever time the specific broker is open. The indicator(s) wer made to reflect accurately trading from start of week, except made to represent one mini lot per pair (so approx. $1/pip) with ability to see pips or $ gains, includes spread, and is the most accurate representation of what Julius' IA was in indicator form IMO.

IF the time of start week needs fixed in indicator, it should be no problem. I have updated code to copy and paste in if needed (if IA doesn't start at beginning of week when broker does, like it should).

Oh yeah, you can look at this old stuff, but shortly I'll be uploading new stuff I had intended to sell for money, but I'm no salesman, and God gives me breath and life freely, takes care of my basic needs, gives me medicine, has allowed me the honor of becoming a coder (well, I get better over time), a Forex trader, an artist, a color scientist, a prepress technician, etc. So I have a love for color and beauty, and enjoy nature He has made. Wow, He really is taking me back to my artistic roots where money didn't matter. Cool. Anything that comes from the life He has given me came to me freely, no matter how hard I work. I could do nothing without Him. I guess I, like Steve Jobs, can say I like waking up and coming to work everyday, for I get to stay at home and work from here. No pay yet, but I wouldn't trade this for going back to corporate world for anything I might have to sell those indicators since it's screw paying any bills and can I afford toilet paper or cigarettes worse. Unemployment just doesn't pay much, but still pays almost as much as if I worked 40 hours a week at a job where they will not pay me near as much as I've made in past, when I'm worth more than that. I just don't know how to make money by donation or open-source, or well, I guess any route really LOL. You need me, and I possibly will need you for this to continue with me coding. I'm really needing God to move mightily, so waiting on Him.

And because a documentary "RevolutionOS" I watched recently (I can watch as many as 4-5 documentaries a night if I don't feel like sleeping, whether coding or not, I still can listen even if not watching) reminded me why open-source came about and that all software used to be freely exchanged between coders before Bill Gates' letter to the community in the 70's, showed me I want to release my software in an open-source way, whether I actually post it as project on SourceForge.net or not. BTW, IF I di put this on SourceForge.net, it looks like it might be the first Forex tools ever posted open-source. That would be cool. But I made my Forex home here on this thread, and unless people want me to leave, I'd rather just stay on this thread and share here. I am committed to beating the brokers, and well, it might just take more than one or two minds collectively to do it I really don't want to have to ask for donations or sell these, but it's getting to the point, what can I do? I don't know. Never been a "money-guy". Just an artist and mathematician with a heart and purpose. Cheers.

Kind regards,

Don

P.S. Even though you may look at these indicators, as I said before, if it doesn't work on single pair, it won't work on basket. They all range at some point or another. Please test the just-given EAs on EURUSD and give feedback if you want to help and no Administrators step forth to try an organize this for us. I'm not an Administrator. I may be many things, but that's not it.

EaglePip,

Thank you

All you should have to do is put the EAs in experts folder, launch MT4 and load template. No need to go into user inputs. If a change is made in user inputs, a new template needs to be made. This is why Administrators need to make new templates, and users need to just load template, tun on expert advisors, and nothing else. Just let it do what it's going to do and report back results at end of week, so that Administrators can keep tally of gains so that this ONE BROKER, ONE PAIR test (we've taken out as many variables as we can by doing this) we can actually compare apples to apples results between different TEMPLATES. One template would shine above the rest is my desire. If something like breakeven is added, new tests would be started.

Every time a variable is changed, we have just made a new set of tests (or doubled current number of tests to be done). This is why if we are to look at this in any type of scientific way, we must limit variables.

How Kos is doing, we'd have to make on M1, both TTF and STF to be M15 in EA (so my EAs are NOT set up as others, because for one, TimeFrame in these indicators are same as TTF, BUT when you call these indicators using iCustom, it's not waiting for the M15 bar to finish, just the M1 bar. This is why this input would need changed to make EA to act like Kos said, because he is not waiting for just the M1 bar to finish, but the M15 bar to finish. I would like to test both ways of each indicator if possible. I've also thought about a 2 EMA test I'd like. I have the ones I want to test already in mind.

OK, so from here on out, if I make a change, I'll try to keep track by putting date and my name at end. If someone else makes a changes, I ask them to do the same, so we all can keep track of when something was done and who did it.

So for me, I did test 20 LWMA waiting on bar close with strategy of M1/M15 agreement to enter and exit. I have found this un-trade-able as is, and have discarded it, and have made both TTF and STF to be M15, and putting it on M15 chart (so basically I have just decided to test equivalent of HAS on M15, no mtf). It took only one night trading and ranging times took more than trend profited. Two sensitive when ranging. So move to M15 is only alternative to adding indicator or break-even code of some sort. In fact I just got trade on M15, and the trade got in like 25 pips too late. So I am discarding LWMA as being too slow entry which market makers are taking advantage. On chart it is clear price swings go only so far as to trigger trade, then reverse. This is not co-incidence. This "ranging" (on purpose swings to screw up indicator traders and non-specific-time traders) kills accounts. HAS and HASmtf have the upper hand on LWMA, so I will now test 2EMAs by myself and see if they come close to being as good as HAS or HASmtf. People may want to try other templates though.

We can add some if someone would like to. My idea is simply (and what I put into Kos EA uploaded, try template with his name) is:
In buy, when low of last bar is higher than entry price and bid is higher than entry price plus one pip, set BE to SL+1 pip. Opposite for sells (high and ask - I might have gotten wrong in his EA and it may need fixed).

Kind regards,

Don

Don& Kos: Hi!

Iam new here, read most of your thread.
Still trying to understand your basket trading system.

Just opened fxdd demo, and started with M15H4_hasmtf_kos.tpl, on eurusd 15 min chart will let you know my results later.

Thank you again both of you for providing this opportunity to learn this concept.
Thank you!
Sincerely,
rswamy

The EAs are not making any trade, surprisingly.

Does that mean without this indicator the EA wouldn't work. Cause all my EAs aren't making any trades.

I had similar experience with the 20 LWMA on the EurUsd 15M.
The EA produced good wins, but needed manual intervention to lock them in. It would reenter after manual exit and had to be suspended to prevent entries in the wrong direction.
However, looking at the code its probably because on a 15m chart it is reading a 300LWMA[20*TTF"15"/STF "1"], which is not very responsive.
However, faster 14Lwma/24LWMA almost perfectly mimic the HAS. It may be a option to use them for quick entry and exits, in a ranging market.

Really like the neat way you have quarantined, the dozens of confusing inputs and simplified the EA.
Will be checking out the VQzz tonight.
EP

Thanks guys for not giving up on the EAs. If having problems, move on to testing Kos version. This version honors Kos' pullback. It may not be perfect yet, I may have to code pullback for only first trade. The pullback code is as follows:
First, agreement: Both HASmtf Timeframe set to 15 and put on M1 chart, and HASmtf Timeframe set to M1 and put on M1 chart, are same color. This sets condition1 to true.
Second, disagreement (pullback): M1now disagrees with M15 HASmtf. Condition2 is now set to true.
Third, agreement again (now resuming trend, and entry time): Now both agree, meaning trend is now resuming, and time to enter, per Kos. Condition3 is set to true.
When all three conditions are true an actual buy has been put on.

NOTE: JUST CAME TO ME. IF NOT HAVING ANY TRADES:
First, make sure that you have smiley face at top right of EURUSD M1 or M15 chart (which you test is up to you since Administrators haven't stepped forth, and I will get to changes as I can and post updates to testing EA(s)). If not, Allow live trading was not checked when you placed EA. Hit F7, go to Common tab, and checkmark Allow live trading, hit OK, and you should see happy face now

Other than that, I know at least one indicator is not working. If an indicator has an arrow, I use the arrow. BUT I am seeing that the arrow "shows after the fact". SO I'll just need to create a for loop like I do to get last top and bottom Semafor 3 each tick, and look over the last x number of bars to see which was the last arrow, and trade accordingly, only allowing one entry per bar (I'm now not allowing exit on that bar, but thinking I want to change that and allow one entry AND exit per bar, so not stuck in bad trade until next bar).

I've also thought of replacing current breakeven code with 3 conditions entry code (will name the duplicate conditions BEconditions), and set to breakeven after a pullback and the Bid is above entry price +1pip in Buy, or Ask is below entry price - 1 pip in Sell.

THEN I've thought about bringing all indicators from one EA, pasting into Kos version, and continue with one EA that obeys pullbacks and we can set BE, no matter which indicator we choose, and the M15 of that indicator will set trend, and the M1 of that indicator will have 3condition entry and breakeven code applied to it. Then we simply can see which one is better, given pullback entry AND BE code in every one of them, since I think by now we would all agree that Kos has given a great feature in BE (and Julius did great by mentioning profit-locking), it's just that now we will use price-action alone with 3conditions to enter, set BE, and set profit-locking after BE if one wants to do so and I get around to coding it.

Having said all that, please also TEST ZZNBRM's EA.

I will be working on EA if I feel like it. Having a hard time this morning, but wanted all of you to know I know there are problems, I want to fix them, and my plans for going forward.

Kind regards,

Don

P.S. EaglePip brings up a good point. I would say that the EA is made to be "modular" by nature, and that like EaglePip, you may see that something you're thinking about would make a better test. The way the EA is designed, there should be AT LEAST ONE "strategy" that is there, whether using iCustom or not, and the code is there for anyone to be able to replace a default strategy with their own, test it, and please give feedback

Also, yes, at this testing stage, if you use my EA, and you choose to use an indicator, that indicator will have to be present in indicators folder. I see not CPU nor RAM advantage while trading one pair, so I'll save Eric some coding and just use iCustom. When testing is done, and if Eric wants to tidy up final EA, that would be greatly appreciated so that an indicator is not needed. But for me, I'm still trying to get entries down, because if I use just colored buffers, then I would be getting in at any time. I'm trying not to enter until NEXT arrow or crossover, etc. That way we don't get in at end of trend by just jumping in ANY time, but ONLY at changes of arrow or color. Which brings the point we could use arrows and not color, color and not arrows, or both on those indicators with arrows. The one dryclean re-introduced to the thread is one that is a default indicator/system, the entries are not right, but if I could use arrows, would like to see these results above all others.

Don: Hi!

Your KOSS system EA, today took number of tades on 1 minute time frame, though they are closing almost instantly.
Thanks!
rs

When I say my multi-indicator/one "system" EA is "modular", I mean:

The variables Trend and TrendColor are global variables, declared at top of EA before init function, whose values can be seen be entire EA and all functions, and which Trend is used to trade stop and reverse strategy, meaning always in an uptrend or downtrend, buy or sell. Waiting on pullback is modification of stop and reverse system, (this particular step is in Kos EA alone right now) implemented by creating more global variables called Conditions, and setting them to true or false as things happen in steps, and when all true, then and only then taking the trade. But I figure we can test all indicators we want, and not just one.

The variables within variable section of EA, where I put my indicators, strategies that define Trend, comments, and draw objects, these are local variables, intended and only really needing to be seen by variables section of code alone.

You see in code something like this:
if(UseVQ7 && !UseVQzz && !UseHASmtf && !UseRHT3 && !Use20LWMAmtf) {

This is saying, in code:
If UseVQ7 is set to true, and UseVQzz is set to false, and UseHASmtf (HASmtf is short for Heiken_Ashi_SmoothedLHm_mtf) is set to false, and UseRHT3 (RHT3 is short for RK-ml-HAMA T3 new - mtf & alert) is set to false, and Use20LWMAmtf is set to false, (so that we are concentrating as we see from all this, on VQ7 alone)...

... and then this is code that anyone could change to change definition of Trend and make EA trade differently.

Just choose one, change if you see that your strategy is better, or you can fix existing strategy (and I mean by strategy, that if I make rules incorrectly, or I did something incorrectly, this is where it will be fixed, under this particular "if" statement, in this "module"), then please do so and give results, and resulting EA with date and your name or username at end.

Kind regards,

Don

rs,

Thanks so much! I will look into that. That of course should not be the case. It's got to either be in "user-intended" code "module" I just described -OR- in actual exit code. At least I know where to look (Just taking a moment to think, it's probably my break-even code that needs changed like I've talked about that is causing this: WAY TO EARLY TO LOCK IN, where 3condition - or 3step - break-even needs to be implemented. I'll look into it)

This makes me think as I work on combo EA, we need to also get Kos EA working correctly, and well, combo-EA working correctly, THEN combine the two.

Thanks again!

Kind regards,

Don

OK. Need to make this clear:

IF USING M15 CHART, DON'T USE M1/M15 TEMPLATE, USE M15/H4 TEMPLATE.
IF USING M1 CHART, DON'T USE M15/H4 TEMPLATE, USE M1/M15 TEMPLATE.

In light of this information above, the code 20*(TTF/STF) yields on M1 chart using M1/M15 template, LWMAs 20 and 300 (20*(15/1)) on M1 chart, NOT M15 chart. On M15 chart, it yields 320 (20*(240/15)) when using M15/H4 template. Close though, so in this regard, whether looking at M1 or M15, close "multiples" are used. I myself have read that multiple of 4 should be used concerning what Kos calls STF and TTF, meaning TTF should be close to multiple of 4 of STF. This leads to using STF/TTF combos like M1/M5, M5/M15, M15/H1, H1/H4, H4/D1, D1/W1, W1/MN1. SEE, in ONE EA, we can test a WHOLE LOT, BUT, FOR TESTING PURPOSES, I FIGURE BEST TO STICK TO M15 and under for STF

Kind regards,

Don

Don: hi!
Thank you for the prompt message.

Yes, seems, breakeven point is too close.
Say, If breakeven point just above the 5min Heiken-ashi, could have made a lot of pips today.
Otherwise, EA entries at 1-15 setup are perfect.

Early morning, I had one buy trade on 15-240 setup as well, closed with breakeven.

Prospects are looking awesome!

Don, thank you again!
Sincerely,
rs

Yes Don, that is correct. Thanks for clarifying.

With the 240/15 template the EA would actually be calculating a 320LWMA ON THE 15M. (Purple line on attached chart)

It definitely looks like BE,Sl and TP are key to a fully automatic system.
In fact just trading the reversals on the 5 LWMA ( when outside the envelope) would capture the major moves, if the fakeouts could be filtered by using BE, SL and TS tied to, say, the ATR.

After gains are locked in with a modest TP, the EA should reenter if indicator still meets entry criteria.
Will attempt to test along those lines.
EP