Discussion re Pure Renko strategy

• 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: 203

  Replies: 8378

  Been thanked: 2024 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: 131

  Replies: 809

  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: 203

  Replies: 8378

  Been thanked: 2024 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.

  Renko Moving Average

  //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"

  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 26 27

  //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"

   

  

  Bard

  Participant

  Topics: 131

  Replies: 809

  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 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"

  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 26 27 28 29 30

  //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"

   

  2 users thanked author for this post.

  eckaw, Paul

  

  Bard

  Participant

  Topics: 131

  Replies: 809

  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: 203

  Replies: 8378

  Been thanked: 2024 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: 203

  Replies: 8378

  Been thanked: 2024 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