• If this is your first visit, you will have to register before you can post. To view messages, please scroll below and select the forum that you would like to visits. Questions? Be sure to check out the Forum FAQ.

Announcement

Collapse
No announcement yet.

Partner 728x90

Collapse

Genetic Optimizer

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

    Genetic Optimizer

    Hi,

    I'm hoping if someone out there who regularly uses genetic optimizer, if they could perhaps share their thoughts and "art of using GO?" I've been reading up a lot on GO and still confused on what they settings should be during my GO tests. Population size, cross rate, generation size etc.

    -ps-n8 support, I did read many of the GO threads and most of them do not have a discussion regarding what are the "best" settings" on "average" to run during a GO test. Perhaps, someone in your data scientist team can also chime in? How did your team come up with the default settings?

    Thank you in advance

    #2
    Hello calhawk01,

    Thanks for your question.

    Genetic backtests factor randomness and evolution to identify ideal parameters instead of running through each and every parameter combination.

    The Genetic Algorithm uses a random set of initial parameters for an initial backtest before it starts generating child parameters that are based off of the performance of these parent parameters.

    Child parameters will be generated from crossovers and mutations of parent parameters. Crossovers will try to find a more ideal parameter between two parent parameters, while mutations add more random variation to the backtest.

    Since we are using random values for the initial values of each backtest and the Genetic Algorithm uses randomness for the mutation of new children and variation, we would expect to see differences between each backtest we run. The purpose behind a Genetic Optimization backtest would be to get a general idea of how fit is for a particular performance metric by using variance and evolution with a smaller number of backtest iterations.

    I would highly recommend to have a look at the Genetic Algorithm video in our publicly available help guide if you have not done so. The video can be a little fast if the Genetic Algorithm concept is a new idea, but it is useful to pause the video periodically to make sure you are up to speed with the presenter.

    This video and further explanations on the Genetic Algorithm can be found in the help guide here: https://ninjatrader.com/support/help..._algorithm.htm

    I would not be able to comment on "best" settings, but maybe other forum members would like to share their experiences. Default parameters would only be chosen to set up a functional Genetic backtest. Changing these parameters will change how the optimization determines new parameters.

    Definitions for these Genetic Optimization parameters can be referenced in our help guide here: https://ninjatrader.com/support/help...ithmParameters

    Please let me know if you have any additional questions.
    JimNinjaTrader Customer Service

    Comment

    Latest Posts

    Collapse

    Topics Statistics Last Post
    Started by rickyblah12, Today, 09:34 AM
    1 response
    4 views
    0 likes
    Last Post NinjaTrader_PatrickG  
    Started by CaptainAmericaXX, Today, 09:05 AM
    2 responses
    10 views
    0 likes
    Last Post koganam
    by koganam
     
    Started by tdawson11, Today, 07:48 AM
    1 response
    6 views
    0 likes
    Last Post NinjaTrader_ChelseaB  
    Started by habibalex, Today, 09:03 AM
    1 response
    5 views
    0 likes
    Last Post NinjaTrader_PatrickG  
    Started by memonolog, Today, 07:56 AM
    1 response
    8 views
    0 likes
    Last Post NinjaTrader_PaulH  
    Working...
    X