Definitely too many parameters. Trying to adjust them for better results will probably lead to overfitting. I also think that’s those stop losses could lead to problems. Possible to do better. Thanks for the post. From a new subscriber.
Just a question. What is your live execution trading platform ?
Totally agree. In the next part we try to make the system more simple. I use real test as my backtesting platform and signal generation and I use interactive brokers for trading
hmm good question. I just tried it too and get the same results as you. I wonder if there was a realtest update that slightly altered how it backtests a bit? I've been looking into it but don't have an answer for you right now
Yea because I had used an older realtest version for a long time and only just upgraded to the latest version. I’d imagine it was a bug fix in real test but I can’t say for sure
Definitely too many parameters. Trying to adjust them for better results will probably lead to overfitting. I also think that’s those stop losses could lead to problems. Possible to do better. Thanks for the post. From a new subscriber.
Just a question. What is your live execution trading platform ?
Totally agree. In the next part we try to make the system more simple. I use real test as my backtesting platform and signal generation and I use interactive brokers for trading
Excited for the second part. I thought exactly the same: too many rules for the entry setup
Thanks Flo! Yes we will definitely have to simplify the entry!
Great post, looking forward to the next one!
Great write up! Looking forward to following along!
Thank you!
Rule3_1: #PercentRank MA(C*V, 20) or Rule3_1: #PercentRank if(inspy, MA(C*V, 20), -9999)
Amazing work! thank you, following with great interest!
Thank you! Glad you liked it
Nicely written :) look forward to part 2
I copied the code you provided at the bottom and pasted it into RT, and my MaxDD value is -12.34% vs -21.13% in your article. How can this be? Thanks
hmm good question. I just tried it too and get the same results as you. I wonder if there was a realtest update that slightly altered how it backtests a bit? I've been looking into it but don't have an answer for you right now
You might be right about RT update. Yet another thing to worry about now, since this is a big difference in the MaxDD value.
Yea because I had used an older realtest version for a long time and only just upgraded to the latest version. I’d imagine it was a bug fix in real test but I can’t say for sure
Thank you
For Realtest software is it necessary to purchase the new license every year ?
If you want the new updates you have to buy a license ever year for 150$ or so. But you can still use the software as is after 1 year.