What I found out in my relative short period as a “trader” that I was not successful. I need to study-Period!
So I found out that the marked seems to repeat it self- that why we, or at least some, have the feeling about where the market will go. Also, markets change and markets are difference. It not the same people that trade the same marked- different psychology and strategy + the fact of FA, yields different price action. So I figured that one strategy is not for every marked. But adjusted it can be fitted. At least when it comes to automated trading. Here we have the ability to sometimes think out if the box and “what If’s?” what if the RSI 14 over bought threshold is not 80, but 78, or the oversold is not 20 but 23. the RSI is in fact better in some other strange period that 14. or the timetables id better from 11:25-18:15 and so on. I believe that the automation can find those combination for you- based on backtesting and optimization. But it’s still up to you to come up with the framework for the strategy. I suggest you look at this video for optimization. This will change everything 🙂 http://www.prorealcode.com/blog/video-tutorials/how-to-optimize-a-trading-system-with-probacktest-prorealtime/
My Proorder codes until now is relying mainly on 4H timeframe but some has 1H and 5min. Until PRT solves the zero bar problem- don’t trust the test that has the profit only in one bar. I mainly test the Pro order code I posted, can’t miss it. It’s the only one I posted 🙂 I have moved to paper trading- and so far- this week it has been profitable, I also do test some codes from the site here that is doing well. I know it’s too soon to tell if the code will be profitable over 3 month, as I only tested it for 2 or 3 weeks. But it is a progression. my first idea was based on Ichimoku. I spend weeks coding only “the holy grail” only to be disappointed by the zero bar profit! you can see why in the screen dump.
I code on Indexes and commodities, but not limited to- I just haven’t got around to other marked with IG. In fact I’m really a newbie when it comes to all the CFD’s, EFT’s, 2.lvl and all those terms. I will learn along the way, when I find a code that I want to try out in the real account. I haven’t found a strategy that fits the forex marked- however there are some nice codes here that I think has potential on Indices. Renko and MM, Donchian Aviator, Lift up and down, DAX 5 min automated, is some that I found here that I think has potential, however I also feel that people will not post all their personal indicators with the code, so their are just basic where you need to spice them up or optimize variables as in the video.
I also think that with multible timeframes in PRT there will be some more interesting ways to go. But I also have a feeling because the code within PRT is so easy to learn it’s very high lvl code and that has limitation. Not that I am very experienced in coding, I’m really not, but I had in the technically studies 20 year ago. It’s just my gut feeling because I read about some other ways to code somewhere on the net. From a perspective that indicators are lagging, they swore to pattern recognition- so they implemented a face recognition code to indentify the patterns. That is heavy, and not something I think can be done with PRT.
Cheers Kasper
.
