Page 1 of 1

Finished

Posted: Sat Sep 05, 2020 9:30 pm
by Jurgen
Just finished the master course.
Very good course, but you lost me on strategy scripts.
I'm now experementing with some own scripts, and I'm gonna start the course allover again to have a more in dept understanding.

Re: Finished

Posted: Thu Sep 10, 2020 5:26 am
by Matthew
Haha that's alright Jurgen, strategy scripts are a whole new ballgame altogether. Mastering indicators and study scripts first will really help you to make better strategy scripts anyway so there's no rush.

Great to have you here mate, I love to work with passionate and motivated traders like yourself. I'm looking forward to seeing what scripts you come up with :)

Re: Finished

Posted: Mon Sep 28, 2020 7:00 pm
by ocaptain
Same here... I'm going to need to learn Strategy Scripts as well as multiple-timeframe analysis... I wrote my first script (thanks to the course!) which I'm excited about but I want to "take it up a notch". Happy coding!

Re: Finished

Posted: Wed Feb 24, 2021 2:52 pm
by dusktrader
Welcome Jurgen. I think strategy scripts are just taking indicator scripts to the next level. You are still analyzing the time series data, but now you can take trading actions as well. This forces a discipline between how the human interprets the indicator and objective actions that you would take (ie, trading plan). For me, the strong focus is that I try to remove subjectivity in my strategy. If I decide a filter is needed, I have to define it in a way that I can automate.

For example currently I'm working on a trend following strategy. I want to catch a chunk of trends, but stay out of choppy markets. As I'm manual trading my rules, I am constantly battling a psychology that wants me to tweak the strategy. I do not tweak my rules unless I can define the tweak in an automate-able way. For example "this bar looks huge" is not quantifiable. It needs to be math-driven. The bar is only huge if I can define that in relative terms. Because that formula is what I need to code. Obviously getting ahead of myself though... I wouldn't code it either until I've proven manually (through manual backtesting) that it is helpful.

As quickly as possible, I would like to learn to code my strategies. I'm going to make a toolbox of common things and then draw from it. This is also how I code in real life. Because much of the effort needed for automated trading is the same across strategies. Once I can assemble a full strategy, then I can leverage the Strategy Tester. I can't wait!!