Short Term Trading Week Starting: 7 Jan, page-102

Hi @Thanky - It's unusual to find someone else who is interested in short term trading system development on HC threads.  Nice to hear from you!  My view re Amibroker hasn't changed since that post from back in August last year.  But you are right about the "rabbit hole" trap.  Although this would apply no matter what systems development package you decided on.

I did eventually decide on a sort term trading system setup focusing on trading DMA CFDs with margin below 30%.  I decided to test it using simple paper trading over about a month prior to doing all the detailed back testing etc work required to complete the system.  When I paper traded the system I found that it did have a good positive edge, and I'm talking about a prospective annual return of at least 50% and hopefully more than 100% on margin.  But I also found that I was opening and closing more that 10 to 15 trades each day and had an ongoing open trade count up to 30 trades on any given day.  For me this made the system unusable, or at the least not suitable for me.  So I had to reassess my proposed system and focus on developing some way to prioritize or rank each trade so I could reduce the open trades list to a smaller figure like 10.  This is what I've been working on over the past few months.  I've been developing ranking indicators that focus on the systems principle components, those being sort term Cycle, short and medium term Momentum, and longer term Trend.  What has surprised me is that the Momentum ranking indicator has turned out to be an indicator that has the potential to be the primary trade setup indicator in the system.  I am now at the stage where I am just about ready to do a new "sanity check" on my revised system.  The big advantage with Amibroker is that it will let you perform this type of portfolio backtesting where individual trades are ranked according to whatever criteria you specify.  You could even use individual stock equity performance over a prior predefined period as a ranking criteria if you wanted to. I'm not aware of any other TA software package that would let you do this. 

Well anyway, I'm not in the situation where I am desperate to have this system in place and generating an income.  Its more a challenge for me to improve the performance of the relatively small amount of money I have in my SMSF.  Its more a hobby for me rather than a business.  So its easy to lose focus at times. Although I am committed to finalizing this short term trading system, and by short term I  mean trades of 1 to 3 and maximum 5 days duration, or ditching the idea completely by the end of this financial year.  The other thing to be aware of is that once you have a system, and due to the fact that time series financial data isn't in any way "stationary", you will need a system for monitoring and assessing the ongoing performance and health of the system, i.e. how do you decide when to modify or even switch off the system.  Howard B Bandy in his book "Quantitative Technical Analysis" presents a statistical based approach using a combined Amibroker and Python based system that I intend to implement and use in the future.  Just another challenge and something more to learn.  When experienced successful Traders tell you that trading isn't as simple or as easy as it would appear to be, you should believe them.  The more of a serious and professional approach you take to your trading the more complicated it becomes and hence the more learning and the more work you have to put into it.  Amibroker, unlike many of the other TA packages, I think will not apply any technological limits to the System Development challenges you'll face.

Good luck in you system development efforts Thanky.  If there is anything I can help you with just let me know, I'd be happy to help if I can.