Discussion re Pure Renko strategy

GraHal · Tue, 25 Feb 2020 10:10:02 +0000

edit : new topic from https://www.prorealcode.com/topic/why-is-backtesting-so-unreliable/ According to IG Status Page … all trading platforms are now sorted!! See attached. This Renko code is the only code I am still getting Rejects / Stoppages on (up to now anyway!?). Only major difference I can see between this System and my others is the use of While, Wend? Maybe While. Wend has recently become a non-approved term anymore?? @Nonetheless would it be easy for you to re-code without While, Wend? I know the code is a DocTrading code so maybe it not easy for you either? If one of our coding wizards might be able to help it would be much appreciated? Below is the code of the System that is showing +£1.8K profit since October (see attached). //------------------------------------------------------------------------- // Main code : Nneless Renko DJI 1M v1 //------------------------------------------------------------------------- //https://www.prorealcode.com/topic/why-is-backtesting-so-unreliable/#post-110889 // Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated // The system will cancel all pending orders and close all positions at 0:00. No new ones will be allowed until after the "FLATBEFORE" time. //DEFPARAM FLATBEFORE = 143000 // Cancel all pending orders and close all positions at the "FLATAFTER" time //DEFPARAM FLATAFTER = 210000 boxSize = 110 //50 once renkoMax = ROUND(close / boxSize) * boxSize once renkoMin = renkoMax - boxSize IF high > renkoMax + boxSize THEN WHILE high > renkoMax + boxSize renkoMax = renkoMax + boxSize renkoMin = renkoMin + boxSize WEND ELSIF low < renkoMin - boxSize THEN WHILE low < renkoMin - boxSize renkoMax = renkoMax - boxSize renkoMin = renkoMin - boxSize WEND ENDIF buy 1 CONTRACT at renkoMax + boxSize stop sellshort 1 CONTRACT at renkoMin - boxSize stop // Stops and targets SET STOP PLOSS 75 SET TARGET PPROFIT 150

This topic has 345 replies, 24 voices, and was last updated 4 years ago by bertrandpinoy.

Tagged: renko

Viewing 15 posts - 286 through 300 (of 346 total)

← 1 2 3 4 5 … 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 →

04/22/2020 at 10:26 AM #127421

GraHal

Bard wrote:

if I can’t live with the % amount of drawdown!

Have you checked out the definition of Drawdown as calculated and shown by PRT?

I’ll try and find my Topic I raised ages ago on Drawdown … it can be misleading?

EDIT / PS

Here we are … have a read of this …

https://www.prorealcode.com/topic/drawdown-useful-or-misleading/

1 user thanked author for this post.

Bard

GraHal

Participant

Topics: 211

Replies: 8761

Been thanked: 2176 times

04/22/2020 at 10:28 AM #127422

Bard

That’s very interesting to hear that there is a difference between the demo and live and that the stops are treated differently, thanks for the heads up Dynstrom. I will have to try it live. I wish PRT would make it easy for users to switch between demo and live and access indicators and systems for both accounts instead of having to go through IG’s platform and switch using “My IG.”

Bard

Participant

Topics: 138

Replies: 814

Been thanked: 99 times

04/23/2020 at 10:20 AM #127579

GraHal

GraHal wrote:

Trailing Stop Type 2 killed it stone dead

You may recall me saying above? To make sure I wasn’t mistaken … I did a comparison check today.

First I duplicated a System to make all else was equal.

Attached shows variable settings for Box size = 120, RenkoType = 3, TrailingStopType = 1 (also = 0!).

Lower curve is with TrailingStopType = 2 … code deleted.

Upper curve is with TrailingStopType = 2 … code included.

My conclusion is that values from TrailingStopType = 2 are being used in TrailingStopType = 1 (and even = o … which should be no TrailingStop!!??) ?

Please can somebody else do tests and post your findings or confirm my findings?

GraHal

Participant

Topics: 211

Replies: 8761

Been thanked: 2176 times

04/23/2020 at 2:31 PM #127616

Paul

yes it’s off! Fifi mentioned it somewhere and I was wrong in my assumption it worked correct.

If you using two trailingstops a fix would be to make every variable different or is the problem where the once command is. I’am 100% sure. That’s the reason I went back to only 1 trailing stop. No fuss.

2 users thanked author for this post.

fifi743,

GraHal

Paul

Participant

Topics: 40

Replies: 812

Been thanked: 660 times

04/24/2020 at 12:38 PM #127788

Bard

Edit:
I got the original graphonprice working, thanks Paul. I also fixed it as a standalone indicator. Cheers.

//moving average
once usema=1
if usema then
once priceMAperiod=14
count=0
sum=0
lastbox=0
for i = 0 to barindex do
if renkomax[i]<>lastbox then
lastbox=renkomax[i]
count=count+1
median=((renkomax[i]+renkomin[i])/2)
sum=sum+median
if count=priceMAperiod+1 then
sum=sum-last
last=median
break
//
endif
endif
next
once avg=undefined
if last<>last[1] then
avg = sum/priceMAperiod
endif
endif
Return avg as "average renko"

//moving average

once usema=1

if usema then

once priceMAperiod=14

count=0

sum=0

lastbox=0

for i = 0 to barindex do

if renkomax[i]<>lastbox then

lastbox=renkomax[i]

count=count+1

median=((renkomax[i]+renkomin[i])/2)

sum=sum+median

if count=priceMAperiod+1 then

sum=sum-last

last=median

break

endif

once avg=undefined

if last<>last[1] then

avg = sum/priceMAperiod

endif

Return avg as "average renko"

Bard

Participant

Topics: 138

Replies: 814

Been thanked: 99 times

04/24/2020 at 12:51 PM #127794

Bard

When you wish there was a delete previous post button… 😀
This is the working Renko Moving Average Indicator:

//Renko moving average
Boxsize = 100
renkoMax = ROUND(close / boxSize) * boxSize
renkoMin = renkoMax - boxSize
once usema=1
if usema then
once priceMAperiod=10
count=0
sum=0
lastbox=0
for i = 0 to barindex do
if renkomax[i]<>lastbox then
lastbox=renkomax[i]
count=count+1
median=((renkomax[i]+renkomin[i])/2)
sum=sum+median
if count=priceMAperiod+1 then
sum=sum-last
last=median
break
//
endif
endif
next
once avg=undefined
if last<>last[1] then
avg = sum/priceMAperiod
endif
endif
Return avg coloured(255,255,0) style(line, 2) as "Renko Moving Average"

//Renko moving average

Boxsize = 100

renkoMax = ROUND(close / boxSize) * boxSize

renkoMin = renkoMax - boxSize

once usema=1

if usema then

once priceMAperiod=10

count=0

sum=0

lastbox=0

for i = 0 to barindex do

if renkomax[i]<>lastbox then

lastbox=renkomax[i]

count=count+1

median=((renkomax[i]+renkomin[i])/2)

sum=sum+median

if count=priceMAperiod+1 then

sum=sum-last

last=median

break

endif

once avg=undefined

if last<>last[1] then

avg = sum/priceMAperiod

endif

Return avg coloured(255,255,0) style(line, 2) as "Renko Moving Average"

2 users thanked author for this post.

eckaw,

Paul

Bard

Participant

Topics: 138

Replies: 814

Been thanked: 99 times

04/24/2020 at 12:56 PM #127796

GraHal

Bard wrote:

When you wish there was a delete previous post button

You can always delete all text and replace with text asking Mods to delete the blank post … if you realise within 5 minutes!? 🙂

1 user thanked author for this post.

Bard

GraHal

Participant

Topics: 211

Replies: 8761

Been thanked: 2176 times

04/24/2020 at 9:09 PM #127921

eckaw

Bard wrote:

When you wish there was a delete previous post button… 😀 This is the working Renko Moving Average Indicator:

Hi @Bard,

I applied this Renko Moving Average code to my renko code and got 100% wins with 4 trades, sl 75 and tp 175 on 1s timeframe. Not sure how it is intended to be used. I used close > avg on long positions and vice verca on short, is that how it is inteded to be used?

eckaw

Participant

Topics: 13

Replies: 118

Been thanked: 61 times

04/24/2020 at 10:08 PM #127926

GraHal

eckaw wrote:

I used close > avg on long positions and vice verca on short, is that how it is inteded to be used?

Do you mean as an extra filter to a Renko System off this Topic?

Feel free to share the full code … we could do with something that works!? 🙂

GraHal

Participant

Topics: 211

Replies: 8761

Been thanked: 2176 times

04/24/2020 at 10:29 PM #127928

eckaw

GraHal wrote:

Feel free to share the full code … we could do with something that works!? 🙂

@GraHal – I don’t know if this works yet, but it looks promising from the backtest – 1s. I’m interested to see how it performs in demo/live with a higher ‘reps’ number, rather than using 2 I’m keen to see how it performs with 8 or more for the analysis of the optimum renko box size. From my observation, I can see that 50 boxsize did really well for a week, but now around 90 boxsize seems to be the right number. So from that analysis I think it doesn’t make sense to ‘optimise’ the renkobox size too often, but rather give it some time to figure out what is the correct setting for the current market movements. I might be wrong but let’s see how it performs next week.

Please note I have have run several renko strategies in backtest in demo that didn’t work out as well as it looked in the backtest. Work in progress!

1 user thanked author for this post.

GraHal

eckaw

Participant

Topics: 13

Replies: 118

Been thanked: 61 times

04/24/2020 at 10:34 PM #127930

eckaw

These are the backtest restulst. I don’t mind less trades per day on this timeframe, as long as the winrate is high. Let’s see how the ML algo works moving forward.I also wonder how the ML code interacts with fixed stoploss and targetprofit code rather than trailing stop and optimised sl. Maybe a fixed value helps in determining the right boxsize?

2 users thanked author for this post.

Paul,

GraHal

eckaw

Participant

Topics: 13

Replies: 118

Been thanked: 61 times

04/25/2020 at 5:30 PM #128044

nonetheless

Hi @eckaw, sadly I’m not getting anything like your results from that itf 🤔

What value are you using for pricemaperiod ?

nonetheless

Participant

Topics: 83

Replies: 1237

Been thanked: 978 times

04/25/2020 at 10:01 PM #128096

eckaw

@nonetheless

I used pricemaperiod = 10 and startingvalue=27

In the backtest the boxsize stayed at 27 as there are only 4 trades. Let’s see what the ML code does on the first optimasation. I will launch on Monday to test and will post the results here.

eckaw

Participant

Topics: 13

Replies: 118

Been thanked: 61 times

04/25/2020 at 10:24 PM #128098

nonetheless

This is what I get, after stripping out the VRT component, which is redundant. Also removed (longonmarket or (not onmarket and tradeon)) and (shortonmarket or (not onmarket and tradeon)) from entry conditions, which belong to VRT.

Although tbh, even leaving all that in, I still couldn’t get anything like your numbers … strange.

nonetheless

Participant

Topics: 83

Replies: 1237

Been thanked: 978 times

04/25/2020 at 11:02 PM #128103

eckaw

I can see the date range is different. I. My backtest I can only access 100000 units in 1s, so 23rd + 24th. You must have 200000. I guess that’s why the results are different, startingvalue 27 was probably not working prior 23rd? I assume your boxsize would have changed from the startingvalue 27 after 8 trades?

eckaw

Participant

Topics: 13

Replies: 118

Been thanked: 61 times

Author

Posts

Viewing 15 posts - 286 through 300 (of 346 total)

← 1 2 3 4 5 … 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 →

Find exclusive trading pro-tools on