Thanks for the feedback. I had written a more detailed with ” ” responses to questions/comments here but lost it because it seems that Chrome doesn’t keep a copy (accept the screenshot attachments) unlike Firefox would if you accidentally closed the tab.
The main point is, if you use dates before the 19th Sept 2018 (which is where we/I thought the tick by tick data started from), with “tick by tick for the backtest” ticked on you still get great tick by tick results that reflect what would have happened intra bar if your entries and exits were activated within the same bar. I rarely use overview screens because they don’t show the drawdown, pls see screenshot. This tick by tick availability with KC is something I have asked IG.
My first thought for Robusta was to set the boxsize to 1/10th of 100, at 10, that didn’t work, neither did 50 but 20 did for a while… Pls see screenshot. Perhaps a good volatility stop like Kaufman’s Volatility Stop or Kase’s Dev Stop should be added?
IG responded to me yesterday to clarify about the 37,500lb contract size: (IG’S commodities spread page it states Coffee Arabica is: “cents/pound $3.75”)
So from the ICE exchange we can see it’s 1 cent per pound. (“Per”? If it was then surely 37500 lbs of coffee would cost a very cheap $375? I think they meant quoted in cents per pound)
Therefore we should do: 375 cents x 100 = 37,500 which is the size of the contract.
Thus a lb of Coffee would cost:
3.75 x 11240.0 at current prices for 37,500 lbs
So if 37,500 (lbs) costs $42,150, you can do $42,150 / 37,500 to get cost / lb which is $1.124.
I also added Vonasi’s Robustness code to the top of Renko ML1 and then got an astronomical “profit” of 240m because of this:
A very important thing when carrying out any robustness testing is to ensure that any money management or increasing and decreasing of position sizing based on profit or loss is turned off.
https://www.prorealcode.com/topic/day-month-year-strategy-robustness-tester/page/4/#post-107553
Does theVRT code look right @GraHal? The rem’d out QYT and Random? I get the results but had a lot of trouble dragging them to this great VRT Excel sheet created by Paul: https://www.prorealcode.com/topic/day-month-year-strategy-robustness-tester/page/7/#post-114611
I did try your one but it said I need to request permission to edit it?
I think to conclude, for current KC trending and ranging markets over the last few years, Renko ML1 with a boxsize =100, works extremely well. Btw I’m not recommending anyone trade with 30 x capital/close.
Next time I see “1M” though I’ll know it means 1 minute and not 1 month… lol, which is the TF I used and how this algo got developed in the first place before being tested on daily, 4 hour and 1 hour all with generally good results, I guess that’s spectral dilation in action.
.
