QuantConnect Community

Going on a tangent: I did find the original list of example algorithms that once upon a time was visible in GUI to be quite useful, that one has since been removed (although I can still get them from

https://github.com/QuantConnect/Lean/tree/master/Algorithm.CSharp

so no loss for me as an experienced user). The example algos could probably be advertised better in GUI (possibly on same page as algo creator when pressing New Algorithm) since these are lower effort on your end than writing a tutorial/bootcamp article. Admittedly, creating a new algo from modules kind of overlaps these, but I think finished examples of how you might write a classical QCAlgorithm still are quite useful to learn how the platform works. Sometimes you just want to do something very simple without using the modules.

I would love a tutorial on how to use multiple insights to make decisions. 

So, this is interesting. Caused me to think of a couple of ideas. I think that the areas that I would like to see in the classes would probably be the module creation and interaction in different areas of the framework. You could run simple strategies through this, that way the users can get a better understanding of how your framework and API operates first hand. For example, you could have a class in alpha model creation, then universe model, and so on. My favorite part of QC is learning how the skeleton and all the different parts work and then pluging in my own ideas and tinkering and messing around with different ideas and strategies. Also, a class on how to create research driven algorithms (stay away from over and under fitting) might be good too. And as a last idea a ML class might be fun. Idk if any of these ideas are good though lol. Just throwing them out there. Can't wait to see what you guys come up with!

Hi Petter,

Glad to hear that you find our demonstration algorithms useful! We value our community's feedback and will take your suggestion into account. Thank you!

1. selection of top 10 shares from SP500 according to a specific creteria e.g. distance to EMA200.
2. statistical evaluation of shooting star candle followed by a certain price drop

Hi Alex,

+1 for Dan proposition.

Also, I think it would be helful to have: (i) a course that covers how to create a custom indicators, and (ii) a course that covers orders management.

Thank you very much!

I would like to see how to use universes. Not only to screen and filter different assets, but also to use them as a "pool" in the algo from which it can pick certain assets or weight them at certain times/under certain conditions.

I would also like to see how to use all the tools and figures shown when backtesting.

Thank you very much!

I've finally got to the place where I understand the order ticketing system when trading a single equity.

Now I'm working on universe selection...which I am finding to be quite complicated. Unsure how to adapt my algorithm and onData to make use of the universe feature. Specifically at this stage I'd like to backtest over the top 50 stocks by dollar volume to get an idea how well my algo performs on different issues.

A tutorial on universes and backtesting over multiple securities, custom plotting -- something a bit more advanced -- would be very helpful to me.

I have noticed that alpha market has not even one alpha for derivatives asset classes like futures or options.  

Bootcamp for derivatives would be very helpful.  For example, how to use futures universes, create alpha and insights.

+1  for Ian Larson and Dan Root propositions

The current Boot Camps are very straight forward. Something more advanced would be nice. Here are some more suggestions for the next Boot Camp Themes:

- Moving Average Strategy, incl. Reports and use of insights for Alpha Market (Level: Easy/Beginner)

- Factor Strategy with automatic reallocation based on some simple portfolio or market conditions (Level: Medium)

- Strategy Portfolio Management: managing several strategies within one Alpha by using an clear object oriented design (Level: Medium/Advanced)

- adressing the single chapters of the LEAN documentation step by step would be a good way imho, especially if combined with some practical applications

Thanks Arthur!

-Futures and Options tutorials. Equities and forex data (due to not expiring) are easy to subscribe to and use, while futures and options have the challenge of selecting liquid contracts/strikes, such as Gold which contracts are active every 3 months. Filtering best practices and guidelines would be useful for these instruments, as well as a runthrough of the API so newer users such as myself can see what's possible

-As Dan said, working on the various elements of the framework and putting them all together would be useful as well!

The thing with having basic bootcamps is...the lvl of API knowledge required to actually make a worthwhile strategy (advanced dealing with selection of assets, orders, etc.) is far more edvanced than the current bootcamps, so really there's nothing but the documentation and forum posts to guide on the journey of actually building something worthwhile. 

Working on it Ian! FYI we've mapped out a path and will be releasing a new BootCamp each week on average following the following path. Sherry Yang is a new team member who will be working on these full time. If there are other contributors who can write at a high level we're also soliciting community contributors to this pathway with a tutorial and compensation laid out here.

1. Buy and Hold with Trailing Stop
2. Momentum-Based Tactical Allocation 
3. Open Range Breakout
4. Liquid Universe Selection
5. 200-50 EMA Momentum Universe
6. Fading The Gap
7. Separation of Concerns with the Algorithm Framework
8. Pairs Trading with SMA 
9. Pairs Trading with Cointegration Test
10. Liquid Value Stocks
11. Sector Balanced Universe Selection
12. Hedging FX Books with Interest Rate
13. Sentiment Analysis of Stocks

Very cool Jared! (Not familiar enough with C# or Javascript, but if I make that change soon enough would be happy to contribute)

extracting complex data from Quandl

Proper order management that covers all important aspects. 

As a bonus, OCO order brackets with triple barrier (SL, PT, EXP) would be very welcome:

SL - Stop Loss
PT - Profit Taker
EXP - Expiration (close after X intervals)

As much as I like working with QC, but building that scenario is surprisingly PITA. 

Big Kudos to the community contributors pathway. Amazing effort and most appreciated! 

Futures tutorial! Building a Futures portfolio for backtesting!

factor analysis and rotation. You got some good machine learning tools but very little documentation and tutorials on how to get the most out of them

A bootcamp on using the documentation, debugger and api in the sidebar would be sensible for beginners!

Hi,

Full BootCamp on consolidator & rollingwindow both applicated on Futures :D

Hey Alex, would great fun to see more Forex related bootcamps. 

Cheers.

Going through the first few bootcamps right now. +1 that api in the sidebar would help (particularlly the QCAlgorithm API). For example figuring out the data type of self.Time was an exercise in frustration with the current docs. 

Suggestion : Using tick data to trade high volume clusters?

Ideas for courses

Swing/Day trade algos
Derivatives, market futures, commodities
Making use of the QC indicators. (pages and pages in the documentation and on github, but no sample implementations).
Avoiding overfitting/selection bias
composite Alphas