Hi Vonasi,
When I make the optimizations of my trading systems I ALWAYS stop in December 2017 for any Time Frame, so the computer cannot know beyond that date what happened …. in the attached graph it starts Without the time-based sale from 1 January 2018. …… clearly I try to have the odds in my favor certainty does not exist in anything …..
I thank you for your work and I wish you a good evening …
PaulParticipant
Master
the set target profit needs to be straight below the buy order because it uses average range & close.
It can not be put on the bottom. It’s placed correct and uses the close when conditions are true to go long.
however, you do not use not longonmarket, so, if your are long and conditions are true to go long again are ignored (because DEFPARAM CumulateOrders = false) , it updates the target with the new close of that bar which is incorrect. To have it correct I think you have to use not longonmarket?
If you add not longonmarket performance goes down, however, if you take every signal when in a long and losing position to close & reenter again long, it’s much better.
here again a but, because not sure why the code below makes such a difference.
if condbuy[1] then
buy positionsize contract at market
SET TARGET PROFIT 5* averagetruerange[1](close)
endif
or
if condbuy[1] then
buy positionsize contract at market
SET TARGET PROFIT 5* averagetruerange[1](close)[1]
endif
// Strategy SR Dow J. Day indicator by roberto gozzi
DEFPARAM CumulateOrders = false
once tradetype = 2 // [1]long & short; [2]long only; [3]short only
once reenter = 0 // [0]off; [1]reenter based on positionperftype
once positionperftype = 0 // [0]perf<>ppv; [1]perf<ppv; [2]perf>ppv
once positionperfvalue = 0
indicator1 = ADX[10]
indicator2 = ADXR[14]
c2 = (indicator1[1] > indicator2[1])
indicator11 = ADX[10]
c1 = (indicator11[1] >= 15)
c30 =TR(close)>25
ONCE PrezzoC = 0
ONCE PrezzoB = 0
ONCE PrezzoA = 0
Prezzo = CALL SRtest
IF Prezzo > 0 THEN
PrezzoA = PrezzoB
PrezzoB = PrezzoC
PrezzoC = Prezzo
ENDIF
positionsize=1
LossMonth =200
once StopMonth=1
IF Month <> Month[1] THEN
ProfitMonth=strategyprofit
StopMonth=1
endif
If strategyprofit<=ProfitMonth-LossMonth then
StopMonth=0
endif
Cond = PrezzoB > PrezzoA AND PrezzoB > PrezzoC
condbuy=StopMonth=1 and close CROSSES UNDER PrezzoB AND C2 AND C1 AND C30 AND Cond > 0
// CONDBUY=CONDBUY AND NOT LONGONMARKET
condsell=1
ctime=1
// entry criteria
if ctime then
if (tradetype=1 or tradetype=2) then
if condbuy then
buy positionsize contract at market
SET TARGET pPROFIT 5* averagetruerange[1](close)
endif
endif
if (tradetype=1 or tradetype=3) then
if condsell and not shortonmarket then
sellshort positionsize contract at market
endif
endif
if reenter then
if positionperftype=1 then
positionperformance=positionperf(0)*100<-positionperfvalue // in loss
elsif positionperftype=2 then
positionperformance=positionperf(0)*100>positionperfvalue // in profit
else
positionperformance=((positionperf(0)*100)<-positionperfvalue or (positionperf(0)*100)>positionperfvalue) // in loss or profit
endif
if (tradetype=1 or tradetype=2) then
if condbuy and longonmarket and positionperformance then
sell at market
endif
if condbuy[1] then
buy positionsize contract at market
SET TARGET PROFIT 5* averagetruerange[1](close) // or [1] at the end?
endif
endif
if (tradetype=1 or tradetype=3) then
if condsell and shortonmarket and positionperformance then
exitshort at market
endif
if condsell[1] and not shortonmarket then
sellshort positionsize contract at market
endif
endif
endif
else
if longonmarket and condsell then
//sell at market
endif
if shortonmarket and condbuy then
//exitshort at market
endif
endif
If longonmarket and (barindex-tradeindex)>=43 then
//SELL AT MARKET
ENDIF
Trailinglong = 2* averagetruerange[5](close)
Trailingsteplong = 5* averagetruerange[1](close)
TGL = Trailinglong
STPL = Trailingsteplong
if not onmarket then
MAXPRICE = 0
PREZZOUSCITA = 0
ENDIF
if longonmarket then
MAXPRICE = MAX(MAXPRICE,close)
if MAXPRICE-tradeprice(1)>=TGL then
PREZZOUSCITA = MAXPRICE-STPL
ENDIF
ENDIF
if onmarket and PREZZOUSCITA>0 then
EXITSHORT AT PREZZOUSCITA STOP
SELL AT PREZZOUSCITA STOP
ENDIF
SET STOP %LOSS 8
however, you do not use not longonmarket, so, if your are long and conditions are true to go long again are ignored (because DEFPARAM CumulateOrders = false) , it updates the target with the new close of that bar which is incorrect. To have it correct I think you have to use not longonmarket?
Well spotted Paul. When I added my SEB testing code to the strategy I encased the STOP and TARGET orders in an IF NOT ONMARKET condition so that I could count all possible trades outside of that condition – so my version did not behave the same as the original. I should try to find time to retest with the varying stop and target orders. Bit busy at the moment as leaving the harbour to go sailing for the summer in a few days time and so lots of jobs to do – including hours and hours of fixing my back-up navigation laptop. Who would think it could be so difficult to get a netbook working on XP!
Hello Paul,
The operation from 1 January 2018 is identical …. and it is the only important thing …… if we had put in real since that date in both ways we would have had the same results.
Thank you
(I am attaching the graph of your variant from January 2018.)
