Is quant and machine learning useful after the test? I made a 35 minute video to showcase my latest results of this forex trading strategy. It uses the combo indicators of moving average cross, statistic number of positive daily returns, and trend. This combo of indicators can be helpful as explained in the video.
NOTE: I have scheduled a webinar to be discussed about this topic. Please find the details below.
It seems that there is a heavy reliance on these for ‘bottoming’ process or testing of flat lining for breakout potential.
Trade like the Pros
I tried to simulate someone like NicTrades later in my video to showcase potential moves with lower trending currency pairs. In short, it seems everything is about perfect timing. You would use all the mentioned indicators as a set of tools to measure timing and historical value of the price. The idea is obviously to capture breakout opportunities for escape volatility return.
It is time to get into the mind of an institutional pro to see how they decide and filter out the trading opportunities. So the struggle continues….. I find NicTrade is a good analyst to watch to see how it is done.
Overlay with fundamentals?
Some will say the fundamentals are not useful. For instance, you could use central bank guidance and trading/econometrics forward looking date to get currency bias on certain countries. I have many scripts ready to overlay. I just need to get a predictable methodology using the technical analysis.
Is Quant and Machine Learning helpful at this point
Some may argue that quant and machine learning techniques may help in this area. I agree but when you do not have a solid foundation, you will even harder times to forecast the markets. I really don’t want to go down these paths of quant or ML is much harder to figure out.
Scheduled webinar info
Hi there,
You are invited to a Zoom webinar.
When: Mar 19, 2018 7:00 PM Eastern Time (US and Canada)
Topic: quant and machine learning useful after this trading backtest
Or iPhone one-tap :
US: +16699006833,,556419907# or +16465588656,,556419907#
Or Telephone:
Dial(for higher quality, dial a number based on your current location):
US: +1 669 900 6833 or +1 646 558 8656
Webinar ID: 556 419 907
International numbers available: https://zoom.us/zoomconference?m=qWa8YsDmCbVsahKNI0p7Ejnd9JYqUR_a
I just read Ivan P’s latest article over at http://www.quintuitive.com/. He mentioned the different backtesting options of Quantopian and QuantConnect. I have no issue with either but I then started think of the different combos of Python and/or R combined. I even thought about 100% Python backtesting framework. Here are a bunch of article below:
I came to the conclusion to keep my Python testing scripts as short and simple as possible. I think what I created through my Python Infrastructure Building Blocks should demonstrate it. Frameworks will only complicate your strategy development workflow.
Another good article from the NYC Conact, he saoid (I think):
If Physics worked like Finance, we would all be levitating since 1687. As soon as Newton would have published his Principia, things would have begun to float around us, and nobody would have been able to verify his equations through experimentation.
