This chart type is deceiving in the NinjaTrader backtester, because the open of the bar is the previous close, which isn't actually what the tick is at. So you get either really great results, or really bad ones, but they aren't how the strategy will actually act.
If you feel inspired, could you please explain why you think they deserve a medal?
Also, could you explain the 'rules' of the Median bar formation.
I would like to write something that can plot the point at which a new bar will be formed in either direction, similar to what I have for Range Bars. I find it very helpful with such bars. The Renkos do seem superior in terms of identifying and sticking with trend moves. But also they can take some wicked whipsaws it seems.
Am I right that these Medians essentially follow the original Renkos but just plot what is happening better versus the current action being missing until new bar is formed?
Or is there something more going on?
PS/Hint: If you feel inspired to code up something like my RangeBarHighLow (in downloads area) for your MedianR's, that would be wunderbar!
I went back in and had another try and this time it seems to have worked although as far as I can tell I am using the same coding. Maybe I had the Custom Type ID nr wrong.
In any case, please check this out to make sure it is correct before I share it in downloads section.
Please register on futures.io to view futures trading content such as post attachment(s), image(s), and screenshot(s).
It seems to be working, but sometimes the plot remains. I don't have this problem with most Range bars but I do with some contracts. Perhaps has something to do with decimals and TickSize calc. But first thing is to ensure the logic is right. Then, since you are such a good programmer, maybe you can ensure that the plots disappear after each bar and don't 'stick' there. But I tink oy gat eet.
Last edited by cclsys; November 24th, 2009 at 06:57 PM.
Reason: replaced code 649pm EST.