"No coding required" solution for algo trading

Discussion in 'Automated Trading' started by fan27, Nov 27, 2019.

Would you consider using the described "no coding required" solution

  1. Yes and I am a programmer.

    6 vote(s)
    12.2%
  2. Yes and I am NOT a programmer.

    10 vote(s)
    20.4%
  3. No and I am a programmer.

    29 vote(s)
    59.2%
  4. No and I am NOT a programmer.

    4 vote(s)
    8.2%
  1. fan27

    fan27

    I have come to a realization that unless you fall into one of the following categories you have absolutely no business venturing into algo trading.

    1. You are able to program.
    2. You have in excess of $10k to get your project off the ground and have a monthly budget for software maintenance costs.

    Based on this, my Strategy Builder is going to be marketed as an fully extendable AlgoTerminal integration that can speed up development time by 10X for those that know how to program. Should have a build within a couple of months for users to try out.
     
    #41     May 19, 2020
    sysdevel99 and qaz like this.
  2. lvca

    lvca

    I still find it easier to write a few lines of Javascript than complicated GUI with pseudo workflows. This is an SMA 50/200 crossover stupid strategy in ArcadeTrader taken from this tutorial:

    upload_2020-6-27_9-41-25.png
     
    #42     Jun 27, 2020
  3. fan27

    fan27

    Yes...if you know how to program, writing code can be preferable to GUI based builders. However, you will find yourself hitting a wall in terms of efficiency by having to code strategies in the matter you are doing. I prefer the best of both worlds which is to write code in a way where I can quickly define new behavior via a config file.
     
    #43     Jun 27, 2020
  4. I found Easylangauge very easy to pick up.

    However, for a more 'flow chart' click and drop approach to algo development, TT ADL is worth a try.
     
    #44     Jun 28, 2020
  5. fan27

    fan27

    Finally going to commercialize this product...should be ready for download within 60 days. The product has come quite far since the original post.
     
    #45     Feb 7, 2021
    userque likes this.
  6. By all means if you can sell even one instance for anything more than free, then I'm your fan.

    In general though ... https://javax0.wordpress.com/2015/12/23/do-not-create-dsl-for-the-customer/

    "Why do most of the DSL projects fail then?

    What I see is that most of these project implement first the switch on the wall and the outlet to plug in the TV, washing machine and so on but then they start to go on and try to invent other tools that would support the reorganization of the wiring. Something like a cross connection table at the electric center of the house where the customer can plug in wires to have 12V direct current in some of the outlets, 120V alternating current in other outlets and 220V in the guest room when visitors come from Europe.

    I hope you get the point: DSL solutions become just too complex for the customers. First it starts off with a simple text file. Later on it is an Excell table. Then a whole bunch of sheets with many columns. Even later somehow scripts emerge that help the editing and the consistency maintenance of the data. And it all supposed to be maintained by the customer.

    The house will burn down! Call the firebrigade!
    "
     
    #46     Feb 7, 2021
    fan27 likes this.
  7. fan27

    fan27

    Totally agree about DSL. I already have customers using the product. Really, I have three potential customers.

    1. Someone who does not want to use the DSL (build config files) and wants everything built for them.

    2. Someone who does not code but wants to use the DSL (build config files) to build their own strategies.

    3. A coder who wants to use the DSL with the ability to add their own customizations.

    I already have customers of type 1 and 2 and I think I have architected in such a way it will be compelling for coders.
     
    #47     Feb 7, 2021
    HobbyTrading and Aquarians like this.
  8. Make sure you know who your audience is. As of now, you asking devs if they would use no-code systems.
     
    #48     Mar 13, 2021
  9. fan27

    fan27

    "No code" is not entirely accurate, hence the reason it was put in quotes. Adding components to the framework definitely requires coding. Using the framework does not. As a developer, if I have to construct each new strategy from scratch it terms of coding, the process is very sub-optimal. What I have built solves that problem.
     
    #49     Mar 14, 2021
  10. A development method/platform has to be complete, no limit, to make any sense. It also need to be fast. Many have tried to solve this problem but so far none have succeeded. Part of the reason is one need to replace programming of everything:

    - of custom indicators since they are input to automated strategies
    - of custom charts since indicators are calculated from charts
    - of custom cross market data


    Further there should be possibility to extend with custom criteria by programming, "possibility" because if above three are implemented, custom criteria would as such not be needed for most.

    There is a platform which has solved all above problems, a Tradestation-class platform. I will post more, but for now a screen shot from the platform showing custom cross market market charts, custom indices and VIX indicator, comparison across symbols etc. Supporting Binance futures, was supporting IB, but will support another similar broker(s). I hope ok to post this picture.

    [​IMG]
    Edit and back-test of a strategy (or portfolio) is done in one click, it takes 200ms when using 100.000 bars.

    The obvious advantages of such 200ms one click functionality, apart from finally being able to move forward with custom automated trading:

    - Learning curve of developing automated strategies has been cut from many years, to days/weeks.
    - Time-to-market for a new set of strategies has been cut from weeks/months to minutes/hours.
    - The above mean one can explore many more types of strategies and ideas, that trading opportunities can be explored and strategies set live within minutes.

    This is a vast subject, where probably the most important to realize is, that any strategy has a time to live, will stop working when market changes behaviour. That market behaviour now changes faster than strategies can be developed by programming. Thereby is strategy development by programming not operationally possible in 2021, as strategies need to be frequently adjusted, even new sets developed.
     
    #50     Aug 22, 2021