Mother of Dragons trading strategy…

nonetheless · Tue, 21 Jan 2020 10:47:18 +0000

Well, I had to call it something, right? Attached is the first code I’ve put together that ‘seems to work’. Survives all manner of WF tests: 70-30, 50-50, 30-70 – however you want to chop it. Also does well in Vonasi’s robustness tester, Random 10-10. I tried hitting it with a very big hammer but still performs well. Uses a combination of HullMA and Supertrend on the 2h TF to define the primary trend, then another HullMA + trailing stop on a 5m TF to control entry and exit. The 5m TF doesn’t give us much in the backtest but it does well in clear uptrends and downtrends. Just that choppy bit in March 2019 where it couldn’t cope. I hate to see whole month where a strategy doesn’t come good, but one out of 16 is not too shabby. It runs around the clock so I’ve allowed for a spread of 3.3, calculated for IG as: (6.5 x 2.4 + 6.5 x 1.6 + 2 x 9.8 + 9 x 3.8)/24 All backtesting was done with MM disabled but I ran the final optimization with it turned on just to demonstrate the boost it gives. Starts with position size = 1 and by the end of the run when equity has trebled, it’s placing position size of ~3.4, so more or less in keeping with standard practise. Also reduces position size in bad patches. Credit where credit is due: Vonasi wrote the Money Management part and coded the HullMA for me. The trailing stop I believe is Nicolas’ work – big thanks to both! But the really cool thing about this strategy is that the equity graph draws a perfect profile of Daenerys Targaryen! You might have to squint a bit, but if you can’t see Daenerys then you haven’t optimized correctly. All comments, suggestions, mods etc welcome! // DJI 5m DEFPARAM CumulateOrders = False // Cumulating positions deactivated DEFPARAM preloadbars = 5000 Capital = 10000 MinSize = 1 //The minimum position size allowed for the instrument. MM1stType = 0 //Starting type of moneymanagement. Set to 0 for level stakes. Set to 1 for increasing stake size as profits increase and decreasing stake size as profits decrease. Set to 2 for increasing stake size as profits increase with stake size never being decreased. MM2ndType = 1 //Type of money management to switch to after TradesQtyForSwitch number of trades and ProfitNeededForSwitch profit has occurred TradesQtyForSwitch = 15 //Quantity of trades required before switching to second money management choice. ProfitNeededForSwitch = 2 //% profit needed before allowing a money management type change to MM2ndType. DrawdownNeededToSwitch = 8 //% draw down from max equity needed before money management type is changed back to MM1stType. DrawdownNeededToQuit = 25 //% draw down from max equity needed to stop strategy Once MoneyManagement = MM1stType Equity = Capital + StrategyProfit maxequity = max(equity,maxequity) if equity < maxequity * (1 - (DrawdownNeededToSwitch/100)) then enoughtrades = 0 tradecount = 0 moneymanagement = MM1stType endif if equity < maxequity * (1 - (DrawdownNeededToQuit/100)) then quit endif if not EnoughTrades then if abs(countofposition) > abs(countofposition[1]) then tradecount = tradecount + 1 endif if tradecount > TradesQtyForSwitch and maxequity >= Capital * (1 + (ProfitNeededForSwitch/100)) then EnoughTrades = 1 MoneyManagement = MM2ndType endif endif IF MoneyManagement = 1 THEN PositionSize = Max(MinSize, Equity * (MinSize/Capital)) ENDIF IF MoneyManagement = 2 THEN PositionSize = Max(LastSize, Equity * (MinSize/Capital)) LastSize = PositionSize ENDIF IF MoneyManagement <> 1 and MoneyManagement <> 2 THEN PositionSize = MinSize ENDIF PositionSize = Round(PositionSize*100) PositionSize = PositionSize/100 // Size of POSITIONS PositionSizeLong = 1 * positionsize PositionSizeShort = 1 * positionsize TIMEFRAME(120 minutes) Period= 520 inner = 2*weightedaverage[round( Period/2)](typicalprice)-weightedaverage[Period](typicalprice) HULLa = weightedaverage[round(sqrt(Period))](inner) c1 = HULLa > HULLa[1] c2 = HULLa < HULLa[1] indicator1 = SuperTrend[5,21] c3 = (close > indicator1) c4 = (close < indicator1) TIMEFRAME(5 minutes) Periodb= 60 innerb = 2*weightedaverage[round( Periodb/2)](typicalprice)-weightedaverage[Periodb](typicalprice) HULLb = weightedaverage[round(sqrt(Periodb))](innerb) c5 = HULLb > HULLb[1] c6 = HULLb < HULLb[1] // Conditions to enter long positions IF c1 AND C3 AND C5 THEN BUY PositionSizeLong CONTRACT AT MARKET SET STOP %LOSS 2.1 SET TARGET %PROFIT 1 ENDIF // Conditions to enter short positions IF c2 AND C4 AND C6 THEN SELLSHORT PositionSizeShort CONTRACT AT MARKET SET STOP %LOSS 1.2 SET TARGET %PROFIT 1 ENDIF //trailing stop function trailingstart = 84 //trailing will start @trailinstart points profit trailingstep = 3 //trailing step to move the "stoploss" //reset the stoploss value IF NOT ONMARKET THEN newSL=0 ENDIF //manage long positions IF LONGONMARKET THEN //first move (breakeven) IF newSL=0 AND close-tradeprice(1)>=trailingstart*pipsize THEN newSL = tradeprice(1)+trailingstep*pipsize ENDIF //next moves IF newSL>0 AND close-newSL>=trailingstep*pipsize THEN newSL = newSL+trailingstep*pipsize ENDIF ENDIF //manage short positions IF SHORTONMARKET THEN //first move (breakeven) IF newSL=0 AND tradeprice(1)-close>=trailingstart*pipsize THEN newSL = tradeprice(1)-trailingstep*pipsize ENDIF //next moves IF newSL>0 AND newSL-close>=trailingstep*pipsize THEN newSL = newSL-trailingstep*pipsize ENDIF ENDIF //stop order to exit the positions IF newSL>0 THEN SELL AT newSL STOP EXITSHORT AT newSL STOP ENDIF //************************************************************************

This topic has 522 replies, 50 voices, and was last updated 3 years ago by LaurentBZH35.

Viewing 15 posts - 211 through 225 (of 523 total)

← 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 … 31 32 33 34 35 →

05/03/2020 at 2:44 PM #129581

eckaw

Hi all, there’s some problems with the code I posted – the heuretics doesn’t work as it should. I will investigate and repost if better results are obtained.

1 user thanked author for this post.

swedshare

eckaw
Participant

Topics: 13
Replies: 118
Been thanked: 61 times

05/03/2020 at 2:45 PM #129582

eckaw

nonetheless wrote:

Nice work with the ML application, are you sure there’s no crosstalk between them?

At the moment there isn’t as the Heuretics code isn’t properly implemented 😉

eckaw
Participant

Topics: 13
Replies: 118
Been thanked: 61 times

05/03/2020 at 2:45 PM #129583

nonetheless

These are the back test results I’m getting – position size 1, MM off.

@Moda, yours is not so far from the original. The results i posted last month were position size 2 to make it easier to compare with the DJ version.

nonetheless
Participant

Topics: 83
Replies: 1237
Been thanked: 978 times

05/03/2020 at 2:55 PM #129591

nonetheless

eckaw wrote:

the Heuretics code isn’t properly implemented

That’s a shame, I tried yours with and without ML and it seemed to improve things. I tried a few ML versions of my own but couldn’t get consistent results so I gave up.

nonetheless
Participant

Topics: 83
Replies: 1237
Been thanked: 978 times

05/03/2020 at 9:34 PM #129644

eckaw

Attached is a version where Heuristics works as it should, and it doesn’t seem to interfere with Paul’s “dynamic live parameter adjustment” https://www.prorealcode.com/topic/dynamic-live-parameter-adjustment/

I’m not sure about the parameters, it looks like the value goes up to maxvalue = 5, which would have increased risk quite a bit. Maybe maxvalue should be lowered?

Results looks promising, will be running this in demo from tomorrow. I used spread 3.8 which is +1 of average spread, as the algo trades 24/7. What are your thoughts on the right spread to use for backtesting this?

eckaw
Participant

Topics: 13
Replies: 118
Been thanked: 61 times

05/03/2020 at 9:34 PM #129647

eckaw

Forgot to attach the .itf! Here it is.

3 users thanked author for this post.

Jabezz, nonetheless, GraHal

eckaw
Participant

Topics: 13
Replies: 118
Been thanked: 61 times

05/03/2020 at 9:38 PM #129650

Francesco

I always use the maximum spread that an asset can reach during the day; i know that it’s a pretty extreme, but if a system can reach good results and shows a good robustness with those spreads, maybe it deserves to run in live.

1 user thanked author for this post.

eckaw

Francesco
Participant

Topics: 32
Replies: 439
Been thanked: 200 times

05/03/2020 at 9:46 PM #129654

eckaw

Francesco wrote:

The ML version is very interesting, also reduces all the noise of the little profit trades. What if applied on the original trailing stop version and not on the less performing breakeven one?

Which version exactly is this @Francesco? I can try to apply it and see what happens.

eckaw
Participant

Topics: 13
Replies: 118
Been thanked: 61 times

05/03/2020 at 9:51 PM #129655

Francesco

Is the original nonentheless’s version on which Moda based his breakeven version. You’ll find it in the attachments of the topic: Dax 5m mother of dragons v2.
The differences with the breakeven versions are not huge, but surely better 🙂
Apart from that, could be interesting to see how ML works on trailing stops, maybe the performances will run up extremely, or maybe could be trash.. who knows

1 user thanked author for this post.

eckaw

Francesco
Participant

Topics: 32
Replies: 439
Been thanked: 200 times

05/03/2020 at 10:37 PM #129683

eckaw

See attached – these are the results I get on the original DAX version with trailing stop. Less total win on Heuristics + “DLPA” (Paul’s Dynamic Live Parameter Adjustment code). Not sure it’s worth it? Parameters could possibly be improved. It wins more but has a bigger max loss.

2 users thanked author for this post.

nonetheless, Francesco

eckaw
Participant

Topics: 13
Replies: 118
Been thanked: 61 times

05/04/2020 at 8:13 AM #129704

MrCool

I still don’t get the same results, looks like it’s only in Januari it’s different. I got 3 losses in a row (see attached picture). Look like none got the same on his test as well.

MrCool
Participant

Topics: 1
Replies: 15
Been thanked: 2 times

05/04/2020 at 9:29 AM #129742

Sebastian Arsjo

I get exactly same as you MRcool

Time settings difference between Sweden and England ?

Sebastian Arsjo
Participant

Topics: 1
Replies: 14
Been thanked: 4 times

05/04/2020 at 11:48 AM #129804

MrCool

Sebastian Arsjo wrote:

I get exactly same as you MRcool

Time settings difference between Sweden and England ?

It is, 1 hour but nonetheless and echaw got different results and they both in England right? Did you get any different results when you moved the time to English time Sebastian?

MrCool
Participant

Topics: 1
Replies: 15
Been thanked: 2 times

05/04/2020 at 3:40 PM #129868

Sebastian Arsjo

No did not try to change, not even sure where i would do that in this new version 🙂

Sebastian Arsjo
Participant

Topics: 1
Replies: 14
Been thanked: 4 times

05/04/2020 at 6:00 PM #129906

MrCool

Sebastian Arsjo wrote:

No did not try to change, not even sure where i would do that in this new version 🙂

Shouldn’t make any difference tho, because there is no time variable in the code so something is wrong. And it’s only Eckaw that is getting this results so?

MrCool
Participant

Topics: 1
Replies: 15
Been thanked: 2 times
Author

Posts