As a new user, it's difficult to have any confidence that the changes I need to make to get them to compile and backtest are even valid. And without working examples, its difficult to learn.
It would be useful, and compelling if the samples would actually ran out of the box, along with parameter sets demonstrating the range of values/condutions under which a range of returns can be expected.
Of the examples I have tried thus far none of them return enough against backtests to even justify a paper trade attempt. Thus far, I have not even been able to find a scalper that demonstrates any real success.
I was very excited about this at first, so I signed up, got interactive brokers account, and started looking into Lean and Algoloop and docker deployment. But without any examples of demonstrable success, and seeing so many algos in the forums begin with optomistic results only to end up with negative returns after their flaws have been ironed out.
I'm starting to wonder if it's even realistic to think this will produce better results than just putting it all into FANG and Tesla.
What am I doing wrong?
Adam W
Do you mean the examples are buggy or simply don't perform very well? If the former, please post the specific bugs either on the forums or open an issue on the LEAN Github.
If the latter, having working examples that are profitable out-of-the-box on backtests can be a bit misleading especially for new users due to potential alpha decay. The examples should be used only as guidelines or simple frameworks to demonstrate how to implement an idea you have in mind, and you could always run your own backtests on them with some parameters to get a sense of the risk-return profile for different classes of strategies.
That being said, most of the examples tend to be some variant of technical analysis (which to be fair, is intuitively appealing and easy to follow for new users). In theory, almost any trading strategy that relies on logic can be done systematically. Why put it all into FANG/Tesla? Are there certain quantifable patterns or factors in them that allow you to infer future prices? Are there other assets with these same patterns?
CAPOCAPITAL
Hang in there, I've struggled myself with the learning curve since my coding abilities are barely competent. I also find myself not always finding the info I'm searching for in the documentation, and have to make sense of things in the github examples, forum questions, and support tickets.
I also joined this space interested in developing scalping algorithms, and have since then changed my approach to be more like active portfolio management of hedged positions. It should be assumed that any algos shared in this space are for educational purposes only, the vast majority of the examples I see aren't shared to demonstrate a profitable system but more to implement a concept using the resources available within QC. No one is going to earn profits in this space cloning another project, it will require developing your own system that is compatible with personal trading style and risk tolerance, and can be decoupled from lucky performance over a backtest. I highly recommend the book Evidence‑Based Technical Analysis, it uses the scientific method to disprove the predictive power of many technical analysis indicators, but teaches a path to test the predictive power of any idea/algo you may come up with using statistical significance tests.
Investing in FANG/TESLA may be the best investment decision we have available today, who knows? Maybe you can develop a strategy that outperforms buy/hold by swing trading during uptrends, or use fundamental analysis to select the cream of the crop. In any case, can we predict the upside potential of FANG/TSLA, or downside risk? It is easy to acknowledge today that these stocks were great picks 10, 15 years ago, but that is hindsight bias. Keep in mind Circuit City was the best performing stock in the 1980's.
Richard Santomauro
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.
To unlock posting to the community forums please complete at least 30% of Boot Camp.
You can continue your Boot Camp training progress from the terminal. We hope to see you in the community soon!