Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Basket Knights - A Basket Trading Round Table 1,310 replies

BASKET TRADE Management: Using multiple trade management EA's on basket trades 8 replies

Updated CCFp and CFP indis for relative strength trading 51 replies

TMS basket: Dashboard for a basket of MA-Filtered RSI Signals 135 replies

CCfp-Diff LIVE Account Trading Journal 50 replies

Attachments: CCFp weekly signals (set and forget basket trading)
Exit Attachments

CCFp weekly signals (set and forget basket trading)

Quoting ajs5274
{quote} Yes I think you can modify anything. Give it a go that's fine.
Well, I added the tabs CCFp 10080, CCFp 1440, CCFp 240, PSM but the issue with the PSM tab I was looking to use is, I would essentially only be able to post my end profit or loss results and I'm not sure how useful that'll be since most of us already believe basket methods can work.

Instead, I'm trying to put together other pair selection methods that can be tested for effectiveness; sort of like a human indicator until the rules are well-defined enough for it to be constructed into an indicator.

I think others might also be interested in finding other pair selection methods. Once it becomes more defined and systematic, they can then forward test and compare their results to estimate what works best to provide the pairs to trade for the basket. But this would put focus on testing different pair selection methods for comparison and I don't know how to organize that editable spreadsheet's PSM tab to make that sort of thing easy for comparison.
Quoting TGS
{quote} Well, I added the tabs CCFp 10080, CCFp 1440, CCFp 240, PSM but the issue with the PSM tab I was looking to use is, I would essentially only be able to post my end profit or loss results and I'm not sure how useful that'll be since most of us already believe basket methods can work. Instead, I'm trying to put together other pair selection methods that can be tested for effectiveness; sort of like a human indicator until the rules are well-defined enough for it to be constructed into an indicator. I think others might also be interested in...
Ok Cheers TGS, thanks for spending the time. Ill have a look, and I am sure it will be very useful.
Hi Guys,

I'm following this thread for a while, when i remember well a few days after start.
The idea was simple, monday morning take the weak and strong pairs based on the indicator.
Open the basket set a sl=200 pips and a tp=500 pips and the money is "comming" based on 3 months data/experience.

We all know by now that it is not that simple.

Then the idea's start popping up. Not start 9 o'clock, let us start 10 o'clock.
We need more baskets, we need different sl, we need different tp.
If we have a profit of 400 pips, let us lock at 200 pips and all its variations.
And when one of us showed a good results for one week, we went into this direction.

Some of us put a lot of time, only think on the poor guy coding 20 different versions of ea with great enthousiasme.

But let us look what is all behind this strategy.
The inputs are the strong and weak pairs based on data of the previous week.
Our input is a currency strength calculation over last week.
We buy pairs from which we know that this pair was last week strong and sell pairs opposite.

But do we have any significant indication that this pair will be also strong or weak in the actual/comming week.
No we dont have. Often a pair which was last week strong is the next week weak or a retrace does happen and becomes strong
again after rewarding us with a nice DD.

Why do we not now.
The price of a pair (currency) is made/driven by let say big money, anyway not by us.
We do not know what price targets the big money has in mind and we do not know how much time they need to reach this
target price and what strategy is behind.
But for sure they say not on monday morning at 8 o'lock or 9 o'clock now we go on with
keeping a pair strong or weak or it is going so well let us extend our startegy for a week. But that is my opinion.
If a pair is strong, weak or ranging can change anytime.
So i think we over estimate the importance of the moment we go into the market.
It can not be true that the success of a strategy depends on the fact if we go at 9 o'clock or at 10 o'clock on monday into the market.

The only assumption we can have is that we expect/hope that the pair which was last week strong or weak will also be
this week strong or weak.
And if most pairs have same behaviour as last week then pairs who change behaviour, we can make some pips.

So if we want to prove if this strategy can work we need to test this over a long period of time, we need statistical significant evidence.
What we can do is take for every week the basket given us by the indicator. Check on an hourly base the max DD for that hour by taking
all H1 low values against open price and look at friday evening what is the result of the basket in pips and what was max DD on hourly base.

If over long time at least two years this brings green pips we have something to discuss.
If we have this on a weekly base we can also check what is the max drawdown in a sequency of weeks compared to max profit in a sequency of weeks.
If then the sun is still shining we can look for improvement.

I know that this is already partly discussed and done in the thread.

For sure i do not want to be negative about this thread (probably sounds a little like that), only i have some points of concern.
I did development work for over 40 years (so be definition I must be an optimist), and my motto is "Even if you fall on your face, you still going forwards"

We jump to my opinion to fast to conclusions based on a few data/evidence and then we go not forwards which leads at the end to frustation
for some people who are working with hart and soul for this thread. For sure i do not want to people them.


Best regards
It is not important how often you fall down, but how often you get up again
Quoting Forexpinbar
{quote} Great job with the EA and updates you create for us on this thread..You are a talented and generous coder..I love you EA..It looks beautiful on the chart and most importantly it does the job pretty well..Good observation about +300 pip move mid week...I think we need to backtest this scenario when the Excel backtest workbook that Excel and his team are working on is completed and published here for public use to help us to verify this scenario and capitalize on it if possible..What do you think?
Sure, when that portion is completed I'll take a dive and determine the probability of the scenario I am looking at.
It is just a simple observation... as we move forward we will see if it is consistent or not. For now, so far it has been consistent... so far.
J
Our AWESOME IMAGINATION is the true holy grail!
Quoting emonts
.... . Some of us put a lot of time, only think on the poor guy coding 20 different versions of ea with great enthousiasme. But let us look what is all behind this strategy. ....... So if we want to prove if this strategy can work we need to test this over a long period of time, we need statistical significant evidence. .......
Folks:
I just got back and saw that my basket was automatically closed by the EA. 32.4% at 806 pips, the high pip count was 894 pips. So I got most of the profit for the week. So did any of you have the EA close out your trades? Especially those of u that have Brokers that close 1 hour early? Please let me know, so we can close this part.

emonts: LOL!! First of all I am not poor. Second we are up to v30, so 30 different versions LOL. 3rd, you hit it right on the nail!! With some time and more data we will be able to calculate a statistical result. I totally agree with that portion. When I analyze the performance of an approach I concentrate on the BAD stuff. Like in this case I look more at the bad weeks, and I look for the statistical edge to GET OUT with a BE or smaller loss. Once that is determined then the other side (winners) are no brainers. IT's the downside that we should concentrate on and get all kinds of numbers and observations. That will then give us the confidence to navigate the market when times are bad and really minimize the loss.
Our AWESOME IMAGINATION is the true holy grail!
Quoting xmess7
When I analyze the performance of an approach I concentrate on the BAD stuff. Like in this case I look more at the bad weeks, and I look for the statistical edge to GET OUT with a BE or smaller loss. Once that is determined then the other side (winners) are no brainers. IT's the downside that we should concentrate on and get all kinds of numbers and observations. That will then give us the confidence to navigate the market when times are bad and really minimize the loss.
Agreed.

Another thing that ones using CCFp-Diff_[v2.0] should consider is creating and organizing data for whether or not the pairs selected, at the time of selection and on the time frame of selection, resulted in a significant move in the predicted direction.

I'm organizing this type of information for a method other than CCFp-Diff_[v2.0] using a spreadsheet with green and red colors and it would be great to see the same sort of data for CCFp-Diff_[v2.0] in case someone has the free time to create such a spreadsheet.
Quoting xmess7
{quote} billsmith: Hi, v18.1 is not setup to handle multiple baskets. I think I mentioned to someone that it was and I was mistaken. PM me and I'll send u a link to one version that does handle multiple basket. Jess
May i have the handle multiple basket version please sir ?
May i know which one is more profitable please sir ?
Thank you very much .
CHECK MY TE !
Davshite All Time Pips: 541
I updated the chart posted midweek with the 5M candles. DD might still be a little underestimated but closer to the possible max DD.

Xmess7, I think you have already added more check points in the newer versions, if not I think they would be very useful (btw many thanks for the work you're doing).

Best pair: EN (all EU pairs ended positive)
Worst pairs: CJ and AJ
AUD pairs caused the largest DD on Tuesday (EA was the worst).
Attached Image (click to enlarge)
Click to Enlarge Name: CCFP_20170306.png Size: 70 KB
Quoting ajs5274
Ok I updated the tradesheet after closing out my trades for the week....smashed it out of the park this week on demo. {image}
Wow! ajjs5274 , -94pips drawdown for 1064pips gain?Thats awesome buddy..Lets hope we have more days like this in future..Profitable trading is about risking less for more.
Goal of a successful trader is to make the best trades. Money is secondary
I'm just looking at this topic real quick and think it would be more time efficient to do things in an automated way.

I made a system to backtest this real quick with the very simple strategy: open at beginning of week, close at end of the week. In the title of the output files you see the date of the week (day.month.year) and the results of that week (pips, number of positions). Then in the file (to be opened with any notepad) you can see the details of the basket (pairs, entry, exit). I'm not absolutely sure if this code is bug free as I made it as a fast example to show you my idea.

I think if you use a similar approach to perform backtests it would be much, much more time efficient. You can of course include basket TP/SL , specific entry/exit times. Draw equity graphs etc etc.

Example for last week:

Report from week starting at 2017.03.05
===================================================================
Buy EURAUD Entry @ 1.39764 Exit @ 1.41454 Result = 169.0 pips
Buy EURGBP Entry @ 0.86302 Exit @ 0.87661 Result = 135.9 pips
Sell GBPCHF Entry @ 1.23877 Exit @ 1.22927 Result = 95.0 pips
Sell GBPJPY Entry @ 139.886 Exit @ 139.563 Result = 32.3 pips
Sell AUDCHF Entry @ 0.76463 Exit @ 0.76134 Result = 32.9 pips
Sell AUDJPY Entry @ 86.406 Exit @ 86.516 Result = -11.0 pips
Sell CADCHF Entry @ 0.75303 Exit @ 0.74919 Result = 38.4 pips
Sell CADJPY Entry @ 85.084 Exit @ 85.174 Result = -9.0 pips
Buy EURCAD Entry @ 1.41942 Exit @ 1.43713 Result = 177.1 pips
Buy EURNZD Entry @ 1.50968 Exit @ 1.54106 Result = 313.8 pips
Sell NZDCHF Entry @ 0.70759 Exit @ 0.69905 Result = 85.4 pips
Sell NZDJPY Entry @ 79.931 Exit @ 79.380 Result = 55.1 pips
===================================================================
Total result this week = 1114.9 pips in 12 positions
Attached Files
File Type: csv 5.2.2017 244.9 pips in 9 positions .csv   < 1 KB | 177 downloads
File Type: csv 5.3.2017 1114.9 pips in 12 positions .csv   < 1 KB | 148 downloads
File Type: csv 12.2.2017 300.0 pips in 9 positions .csv   < 1 KB | 127 downloads
File Type: csv 19.2.2017 386.2 pips in 16 positions .csv   1 KB | 156 downloads
File Type: csv 22.1.2017 707.9 pips in 12 positions .csv   < 1 KB | 136 downloads
File Type: csv 26.2.2017 -432.0 pips in 8 positions .csv   < 1 KB | 141 downloads
File Type: csv 29.1.2017 -1336.3 pips in 8 positions .csv   < 1 KB | 148 downloads
What if the EA is made to trade using both weekly and hourly strength consideration of indicator ccfp to reduce the effect of difference in start up run time. Also this will ensure the EA weekly signal is confirmed by hourly signal thus increase probability of profit.
Know News/time; WAIT for Edge Entries, MPLC Normalization, Exits & TP's..
Quoting L.m
{quote} Great job Tiivaa. Here are the updated tables. @Forexpinbar Since Ficibald already tested the 50 pips SL, there's no need for me to do it. About the 50 days ATR actually I think it's too big, a 15 pairs basket would have +2000 pips SL and it doesn't make sense to me. Remember that we are talking about basket SL and TP, it would be different if you wanted to set individual SL and maybe also TP per every single pair. {image}
Hi L.m , can you run a backtest for 2016 from January to December and share with us now that we have the past weekly buy and signals that was compiled by Smoothtrader using the the same TP and SL criteria you used in this excel result you compiled? I am still trying to figure out how to run the backtest in Excel lol...though this backtest may not be accurate , it will be worth looking at....When i complete my learning on how to run backtests in Excel i will run backtests and share results..Hope you consider my request
Goal of a successful trader is to make the best trades. Money is secondary
Hi all,

Herewith the results of my demo trading this week, 6 till 11 march 2017. I have the results for 3 tests, used V27 of the EA.
1) Auto start last monday at 09 GMT, Basket SL200/Basket TP500. No individual TP/SL. Basket SL (200) hit at 09:39GMT.
(sorry, i don't have a screenshot of this). -200pips

2) Entered again the same weekly basket at 10 GMT, now Basket SL 500/Basket TP500. Wednesday 8/3 at 09:37GMT TP (500) was hit and all trades automatically closed. +500pips

3) Basket start 09 GMT, this with no basket tp/sl but with Individual TP/SL (500/200). No individual tp/sl was hit and basket was automatic closed friday evening 20mins before market closed. +161 pips.

Many thanks to ForexPinbar and all contributors of this forum for developping and improving this strategy and for coding the EA!!!

Piet.

p.s. can anyone explain me please how i can modify or show me the way how to use the indicator CCFp_Diff_[v2.0](1).ex4 to show me the pair selection results from an long time ago previous week e.g. KW10 last year? thanks!
Attached Images (click to enlarge)
Click to Enlarge Name: Results FXCM demo.JPG Size: 99 KB Click to Enlarge Name: Results Intertrader demo KW10 2017.JPG Size: 106 KB
Quoting scalprobot
I'm just looking at this topic real quick and think it would be more time efficient to do things in an automated way. I made a system to backtest this real quick with the very simple strategy: open at beginning of week, close at end of the week. In the title of the output files you see the date of the week (day.month.year) and the results of that week (pips, number of positions). Then in the file (to be opened with any notepad) you can see the details of the basket (pairs, entry, exit). I'm not absolutely sure if this code is bug free as I made...
Thanks for your efforts to post results.

I assume only one pair at a time can be backtested using your simulation system and if so, each pair would have to be run with a small starting deposit relative to the lot size so that its drawdown during that week could be determined.

For example, the Buy EURAUD Entry @ 1.39764 Exit @ 1.41454 Result = 169.0 pips, resulted in 169 net pips but what drawdown did it experience during that week to ultimately finish at 169 net pips?

It's be good to see the end week results of the simulation system for each of the above pairs using 0.10 lot and let's say a $500 account. I must say though that even after seeing those results if you were to produce them, it still leaves a lot of questions unanswered because this is attempting to back test a basket of trades and obviously there are software limitations with MT4 for backtesting this type of thing; maybe it's different with MT5?
hey bro. can u put the image of the backtest in ur EA? i donnot know how to test it . thanks
Quoting forexmercuri
Hi all, Herewith the results of my demo trading this week, 6 till 11 march 2017. I have the results for 3 tests, used V27 of the EA. 1) Auto start last monday at 09 GMT, Basket SL200/Basket TP500. No individual TP/SL. Basket SL (200) hit at 09:39GMT. (sorry, i don't have a screenshot of this). -200pips 2) Entered again the same weekly basket at 10 GMT, now Basket SL 500/Basket TP500. Wednesday 8/3 at 09:37GMT TP (500) was hit and all trades automatically closed. +500pips 3) Basket start 09 GMT, this with no basket tp/sl but with Individual TP/SL...
You can download all the past CCFp weekly signals you need from here https://www.forexfactory.com/showthr...07#post9644607 ...Signals were compiled by Smoothtrader from 2012 to 2017..All thanks to Smoothtrader for doing a great job..
Goal of a successful trader is to make the best trades. Money is secondary
Attached Image
Goal of a successful trader is to make the best trades. Money is secondary
Quoting aimsttrader
{quote} May i have the handle multiple basket version please sir ? May i know which one is more profitable please sir ? Thank you very much .
The latest beta version is v30, it handles multiple baskets (make sure u have different magic numbers).
You may download it at: http://xm7.info/ea/ccfp_weekly/
NOTE: Since it is beta you may experience errors/bugs PLEASE PM me with any problems so we can solve them away from the thread

Which is more profitable? That is up to you to find out.
All I can say is that we really have a winner, it is just a matter now of figuring out the stats for the losing weeks so we can know WHEN to get out and minimize any loss
Our AWESOME IMAGINATION is the true holy grail!
GMT hour

Folks:
Some of you are making it too complicated. It's really quite simple.
GMT is GMT no matter when....

For example on pg1 the author indicates:
5.Every Monday exactly when London opens , enter all the signals you see in the CCFp indicator with your broker(demo account) at the same time London opens.

London open is defined as 08:00. So the inputs to the EA are:
OpenGMTHour="08:00"
_GMTDayOfWeek_ = Monday

If I say for example open basket on Wednesday at 17:00 GMT then enter those very numbers in EA
OpenGMTHour="17:00"
_GMTDayOfWeek_ = Wednesday


Ok I understand that some of you want to interpret the open hour using your time or your broker's time.
All u need to do is one step, calculate the GMT time of your time or your broker's. That is it, then u just enter those very numbers into the EA.

In my case I live in GMT-8 (Pacific Time). So if I want my basket to open at my time around 18:00 Thursday.
I would first calculate the GMT time like this.
18:00 + 8 hrs (cause I am behind GMT by 8) = 02:00 Friday
So in the EA I then enter :
OpenGMTHour="02:00"
_GMTDayOfWeek_ = Friday
NOTE: Why Friday?? Cause when I added 8 to 18 I actually when beyond midnight into the next day.

So in summary, just keep it simple. If you have a doubt just ask someone that knows, and just get the GMT time and GMT day.. That is all u need to know.
Our AWESOME IMAGINATION is the true holy grail!
Quoting L.m
I updated the chart posted midweek with the 5M candles. DD might still be a little underestimated but closer to the possible max DD. Xmess7, I think you have already added more check points in the newer versions, if not I think they would be very useful (btw many thanks for the work you're doing). Best pair: EN (all EU pairs ended positive) Worst pairs: CJ and AJ AUD pairs caused the largest DD on Tuesday (EA was the worst). {image}
L.m:
I was thinking about the same thing. In my case I am trying to control the cpu/memory imprint of the EA. So at first I limited the EA to recording the pip total once every hour. But then I saw that the EA was not loading the system that badly (3% CPU and about 450Megs of RAM), so in the latest version I raised the resolution to once every 15min. I think the graphs that I post give the impression that more points are recorded. BTW those graphs are just me opening up the csv file in excel then automatically applying one of the graph templates. I get rid of some data points that aren't needed and then just take an image of the resulting graph.
I am considering putting a button on the chart that when clicked will automatically popup a chart similar to what I posted. That way I would not get users asking "Where is the file located?" LOL

As you can imagine there are many more things we can do with the EA when it comes to plotting this stuff. But until we have that last bug free EA, that will have to wait.

Peace
Our AWESOME IMAGINATION is the true holy grail!
1 trader viewing now