Money management & control risk

deleted271219 · Thu, 15 Aug 2019 20:19:12 +0000

Anyone know how to code Kelly’s F Optima for PRT ??? Thanks, is very important for a good medium por long strategie.

This topic has 23 replies, 4 voices, and was last updated 5 years ago by Vonasi.

Tagged: Kelly’s optimal F, money management

Viewing 9 posts - 16 through 24 (of 24 total)

← 1 2

08/20/2019 at 4:24 PM #105117

Vonasi

I found this EasyLanguage calculator of F in the September 26 2018 post of this blog:

http://georgepruitt.com/category/easylanguage/must-know/

I know nothing of EasyLanguage coding but I see that it uses arrays which may make conversion to PRT either impossible or very difficult.

Perhaps Nicolas could comment on whether it could be converted?

Note: DO NOT try to use the code posted here in PRT as it is – it is EasyLanguage code!

//OptimalFGeo by George Pruitt
//My interpretation Sept. 2018
//www.georgepruitt.com
//georgeppruitt@gmail.com

input: minNumTrades(numericSimple);
vars: totalTradesCount(0),tradeCnt(0);
array: tradesArray[500](0);

vars: iCnt(00),jCnt(00),grandTot(0),highI(0);
vars: optF(0.0),gMean(0.0),fVal(0.0),HPR(0.0),TWR(0.0),hiTWR(0.0);
vars: biggestLoser(0.0),gat(0.0);

totalTradesCount = totalTrades;
If totalTradesCount > totalTradesCount[1] then
begin
	tradeCnt = tradeCnt + 1; 
	tradesArray[tradeCnt] = positionProfit(1);
end;

// Taken from my Fortran library - GPP and Vince Book PMF

optF = 0.0;
gMean = 1.00;
gat   = 0.00;
//Only calculate if new trade
IF(tradeCnt>minNumTrades and totalTradesCount > totalTradesCount[1]) then 
Begin
	biggestLoser = 0;
	grandTot = 0;
	For iCnt = 1 to tradeCnt //get the biggest loser
	begin
   		grandTot = grandTot + tradesArray[iCnt];
   		IF(tradesArray[iCnt]<biggestLoser) then biggestLoser = tradesArray[iCnt];
	end;
//	print(grandTot," ",biggestLoser);
	IF({grandTot > 0 and} biggestLoser <0) then 
	begin
//		print("Inside TWR Calculations");
		highI = 0;
		hiTWR = 0.0;
		for iCnt = 1 to 100
		begin
			fVal = .01 * iCnt;
			TWR = 1.0;
			for jCnt = 1 to tradeCnt // calculate the Terminal Wealth Relative
			begin
    			HPR = 1. + (fVal * (-1*tradesArray[jCnt]) / biggestLoser);
    			TWR = TWR * HPR;
 //   			print(fVal," ",iCnt," " ,jCnt," Trades ",tradesArray[jCnt]," HPR ",HPR:6:4," TWR : ",TWR:6:4," hiTWR",hiTWR:6:4," bl ",biggestLoser);
			end;
//			print(iCnt," ",TWR," ",hiTWR);
			IF(TWR>hiTWR) THEN
			begin
    			hiTWR = TWR;
    			optF = fVal;    	// assign optF to fVal in case its the correct one		
			end
			else
    			break;                     //highest f found - stop looping
		end;		
		If (TWR <= hiTWR or optF >= 0.999999) then
		begin
			TWR  = hiTWR;
			OptimalFGeo = optF;  //assign optF to the name of the function
		end;	
		gmean = power(TWR,(1.0 / tradeCnt));
		
		if(optF<>0) then GAT   = (gMean - 1.0) * (biggestLoser / -(optF));		
		print(d," gmean ",gmean:6:4," ",GAT:6:4);  // I calculate the GMEAN and GeoAvgTrade
	end;
end;

//OptimalFGeo by George Pruitt

//My interpretation Sept. 2018

//www.georgepruitt.com

//georgeppruitt@gmail.com

input: minNumTrades(numericSimple);

vars: totalTradesCount(0),tradeCnt(0);

array: tradesArray[500](0);

vars: iCnt(00),jCnt(00),grandTot(0),highI(0);

vars: optF(0.0),gMean(0.0),fVal(0.0),HPR(0.0),TWR(0.0),hiTWR(0.0);

vars: biggestLoser(0.0),gat(0.0);

totalTradesCount = totalTrades;

If totalTradesCount > totalTradesCount[1] then

begin

tradeCnt = tradeCnt + 1;

tradesArray[tradeCnt] = positionProfit(1);

end;

// Taken from my Fortran library - GPP and Vince Book PMF

optF = 0.0;

gMean = 1.00;

gat = 0.00;

//Only calculate if new trade

IF(tradeCnt>minNumTrades and totalTradesCount > totalTradesCount[1]) then

Begin

biggestLoser = 0;

grandTot = 0;

For iCnt = 1 to tradeCnt //get the biggest loser

begin

grandTot = grandTot + tradesArray[iCnt];

IF(tradesArray[iCnt]<biggestLoser) then biggestLoser = tradesArray[iCnt];

end;

// print(grandTot," ",biggestLoser);

IF({grandTot > 0 and} biggestLoser <0) then

begin

// print("Inside TWR Calculations");

highI = 0;

hiTWR = 0.0;

for iCnt = 1 to 100

begin

fVal = .01 * iCnt;

TWR = 1.0;

for jCnt = 1 to tradeCnt // calculate the Terminal Wealth Relative

begin

HPR = 1. + (fVal * (-1*tradesArray[jCnt]) / biggestLoser);

TWR = TWR * HPR;

// print(fVal," ",iCnt," " ,jCnt," Trades ",tradesArray[jCnt]," HPR ",HPR:6:4," TWR : ",TWR:6:4," hiTWR",hiTWR:6:4," bl ",biggestLoser);

end;

// print(iCnt," ",TWR," ",hiTWR);

IF(TWR>hiTWR) THEN

begin

hiTWR = TWR;

optF = fVal; // assign optF to fVal in case its the correct one

end

else

break; //highest f found - stop looping

end;

If (TWR <= hiTWR or optF >= 0.999999) then

begin

TWR = hiTWR;

OptimalFGeo = optF; //assign optF to the name of the function

end;

gmean = power(TWR,(1.0 / tradeCnt));

if(optF<>0) then GAT = (gMean - 1.0) * (biggestLoser / -(optF));

print(d," gmean ",gmean:6:4," ",GAT:6:4); // I calculate the GMEAN and GeoAvgTrade

end;

Vonasi

Moderator

Topics: 192

Replies: 5088

Been thanked: 1811 times

08/23/2019 at 1:31 PM #105384

Nicolas

This is a rough version, it makes the computation but i used a positionprofit in percentage instead of real money.. so I assume the result is not really the one that we need. The problem is to know the position profit in money by using positionperf at a time when we don’t know what was the Close value used to calculate it.

defparam cumulateorders=false

// --- settings
minNumTrades = 10
// --- end of settings

//dummy strategy
irsi = rsi[14]
if irsi crosses over 50 then
buy at market
tradeCnt = tradeCnt+1
endif
set target pprofit 20
set stop ploss 10

//
optF = 0.0
gMean = 1.00
gat = 0.00
IF(tradeCnt>minNumTrades and tradeCnt<>tradeCnt[1]) then
biggestLoser = 0
grandTot = 0
For iCnt = 1 to tradeCnt do //get the biggest loser
grandTot = grandTot + positionperf(iCnt)
IF(positionperf(iCnt)<biggestLoser) then
biggestLoser = positionperf(iCnt)
endif
next
IF(grandTot > 0 and biggestLoser <0) then
//highI = 0
hiTWR = 0.0
for iCnt = 1 to 100 do
fVal = .01 * iCnt
TWR = 1.0
for jCnt = 1 to tradeCnt do // calculate the Terminal Wealth Relative
HPR = 1.0 + (fVal * (-1*positionperf(jCnt) / biggestLoser))
TWR = TWR * HPR

IF(TWR>hiTWR) THEN
hiTWR = TWR
optF = fVal // assign optF to fVal in case its the correct one
else
break
endif
next
next //highest f found - stop looping
If (TWR <= hiTWR or optF >= 0.999999) then
TWR = hiTWR
OptimalFGeo = optF //assign optF to the name of the function
endif
//gmean = power(TWR,(1.0 / tradeCnt))
//y = x^p = e^(p*LN(x))
//This could then be coded as: y = EXP(p*LOG(x)) This works fine.
gmean = exp((1.0/tradeCnt)*log(TWR))

if(optF<>0) then
GAT = (gMean - 1.0) * (biggestLoser / -(optF))
//print(d," gmean ",gmean:6:4," ",GAT:6:4); // I calculate the GMEAN and GeoAvgTrade
endif
endif
endif

graph gat
graph OptimalFGeo
//graph tradeCnt
//graph positionperf

defparam cumulateorders=false

// --- settings

minNumTrades = 10

// --- end of settings

//dummy strategy

irsi = rsi[14]

if irsi crosses over 50 then

buy at market

tradeCnt = tradeCnt+1

endif

set target pprofit 20

set stop ploss 10

optF = 0.0

gMean = 1.00

gat = 0.00

IF(tradeCnt>minNumTrades and tradeCnt<>tradeCnt[1]) then

biggestLoser = 0

grandTot = 0

For iCnt = 1 to tradeCnt do //get the biggest loser

grandTot = grandTot + positionperf(iCnt)

IF(positionperf(iCnt)<biggestLoser) then

biggestLoser = positionperf(iCnt)

endif

IF(grandTot > 0 and biggestLoser <0) then

//highI = 0

hiTWR = 0.0

for iCnt = 1 to 100 do

fVal = .01 * iCnt

TWR = 1.0

for jCnt = 1 to tradeCnt do // calculate the Terminal Wealth Relative

HPR = 1.0 + (fVal * (-1*positionperf(jCnt) / biggestLoser))

TWR = TWR * HPR

IF(TWR>hiTWR) THEN

hiTWR = TWR

optF = fVal // assign optF to fVal in case its the correct one

else

break

endif

next //highest f found - stop looping

If (TWR <= hiTWR or optF >= 0.999999) then

TWR = hiTWR

OptimalFGeo = optF //assign optF to the name of the function

endif

//gmean = power(TWR,(1.0 / tradeCnt))

//y = x^p = e^(p*LN(x))

//This could then be coded as: y = EXP(p*LOG(x)) This works fine.

gmean = exp((1.0/tradeCnt)*log(TWR))

if(optF<>0) then

GAT = (gMean - 1.0) * (biggestLoser / -(optF))

//print(d," gmean ",gmean:6:4," ",GAT:6:4); // I calculate the GMEAN and GeoAvgTrade

endif

graph gat

graph OptimalFGeo

//graph tradeCnt

//graph positionperf

1 user thanked author for this post.

Vonasi

Nicolas

Keymaster

Topics: 81

Replies: 20863

Been thanked: 4774 times

08/23/2019 at 3:26 PM #105400

Vonasi

I think it will take a few moments to get my head round what that code is trying to do. Perhaps it would be better to try to code it as an indicator with a simulated strategy built in rather than a strategy?

Vonasi

Moderator

Topics: 192

Replies: 5088

Been thanked: 1811 times

08/24/2019 at 5:30 PM #105454

deleted271219

No, I can’t … the code is yet to be completed.

A few days ago I asked how to enter Kelly’s F or Kelly’s Secure F as money management in the code and I still haven’t got an answer.

In addition, I want to get a couple of things for the code:

For each successful operation add 10% more risk than in the previous operation for the next operation, and for each failed operation remove 10% less risk than in the previous operation.
If the total benefit of the system is for example between € 10,000 or € 12,000, operate with € 5 / pip. If the total benefit of the system falls, and is between € 8,000 or € 10,000 operate with € 4 / pip. But if then the total benefit of the system is once again between € 8,000 or € 10,000, then re-operate with € 5 / pip. With these ideas the code would be complete and could upload it.

deleted271219

Member

Topics: 3

Replies: 21

Been thanked: 0 times

08/24/2019 at 5:39 PM #105456

robertogozzi

@sdesergio

do NOT use Insert PRT code for TEXT, it doesn’t work!

In you last request did you refer to the current discussion or to Kelly’s Secure F?

robertogozzi

Moderator

Topics: 69

Replies: 17188

Been thanked: 4064 times

08/24/2019 at 7:47 PM #105465

deleted271219

<span lang="en" tabindex="0">Let's see, all I say is that the code is unfinished and I need help.
Also in September I will have 200k of backtest and I will have to update it very probably.

Until then it is nonsense to upload a half-programmed code.

I don't force anyone to "work for me" ... I've only asked for help just like others like me do.

Im referred F of Kelly and actually discusión.</span>

<span lang="en" tabindex="0">Let's see, all I say is that the code is unfinished and I need help.

Also in September I will have 200k of backtest and I will have to update it very probably.

Until then it is nonsense to upload a half-programmed code.

I don't force anyone to "work for me" ... I've only asked for help just like others like me do.

Im referred F of Kelly and actually discusión.</span>

deleted271219

Member

Topics: 3

Replies: 21

Been thanked: 0 times

08/24/2019 at 9:22 PM #105468

robertogozzi

@sdesergio

do NOT use Insert PRT code for TEXT, it doesn’t work!

robertogozzi

Moderator

Topics: 69

Replies: 17188

Been thanked: 4064 times

08/24/2019 at 9:27 PM #105471

robertogozzi

@sdesergio

don’t mix post from different topics
try not to urge help, IF someone is able AND willing to answer… he will, provided you have posted all details and answered any question may have been asked

urging people to work for you when you want them to… does not work as intended!

robertogozzi

Moderator

Topics: 69

Replies: 17188

Been thanked: 4064 times

08/24/2019 at 10:17 PM #105479

Vonasi

sdesergio – no one is working here. No one is employed or paid by anyone else to provide free coding solutions to anyone else. This is a forum set up by someone who is not employed by PRT just to provide a community for those who use PRT. It is moderated by volunteers.

The Optimal F calculation that you request has been discussed in the past and no one was able to provide the code then. We are now a tiny step closer thanks to the Easylanguage code that I found and thanks to Nicolas’s conversion of it- all this despite a lack of actual information on how exactly Optimal F is calculated from yourself or from any sources found by anyone else and yet you are complaining that no one has provided you with an answer. Then you throw in a whole bunch of new demands on top!

As moderators we have had to constantly remind you of basic forum etiquette and to tidy up your posts after you and as volunteers giving our time for free we would appreciate it if you took up slightly less of our time doing this.

At the moment your posts are likely to get you less help rather than more help. 🙂

Vonasi

Moderator

Topics: 192

Replies: 5088

Been thanked: 1811 times

Author

Posts

Viewing 9 posts - 16 through 24 (of 24 total)

← 1 2

Find exclusive trading pro-tools on