The Power of AI in Trading: Programming your first Machine learning model to predict next candle.

Поделиться
HTML-код
  • Опубликовано: 22 янв 2025

Комментарии • 22

  • @skh930
    @skh930 7 месяцев назад +3

    One of the best channel I've ever seen, good job

  • @samankhavandkary5927
    @samankhavandkary5927 7 месяцев назад

    We were waiting for your new video 👍

  • @نیانلیلیپور
    @نیانلیلیپور 6 месяцев назад

    vielen dank fur ihre praktische informationen

  • @shamanaggarwal4696
    @shamanaggarwal4696 5 месяцев назад

    yes pls cover the topics like variance and covariance and Rsquare

    • @ecoengineering7535
      @ecoengineering7535  4 месяца назад

      Hi, thank you very much for your suggestion, I will.

  • @aljilanihadist5955
    @aljilanihadist5955 Месяц назад

    i know this is 5 months ago and i dont know if you check comments. my question after greeting you with respect. i downloaded ollama trading model how can we use it to let it trade in simulation i made python gui to submit posts to it. cheers

    • @ecoengineering7535
      @ecoengineering7535  24 дня назад

      Hi, I am not sure about your question... can you say it more clearly, thank you

  • @تیپیتریدر
    @تیپیتریدر 6 месяцев назад

    سلام تشکر از کانال بسیار عالی شما.من استراتژی ار اس ای معکوس شما رو دیدم و خیلی ازش خوشم اومد رفتم زبان MQL5 رو یاد گرفتم و ربات رو طبق اون چیزی که شما گفته بودید نوشتم اما نتایجم مثل شما نبود میشه لطفاً راهنماییم کنید ممنونم 😊

  • @Jabir_AfreDy
    @Jabir_AfreDy 6 месяцев назад

    Can you develop hedging ea please

    • @ecoengineering7535
      @ecoengineering7535  4 месяца назад

      we have already implemented it in our channel. Please find it here:
      ruclips.net/video/3dAs980enns/видео.html

  • @robertbyer2383
    @robertbyer2383 7 месяцев назад +1

    Great video, but I think there are a few issues.
    The major issue is how you're splitting the data in to training and test sets. Since you are working with time based data using the built in "split" functions to split the data will cause inaccurate training because the split function will take "random" samples which means that some of the "training" data points could be in a time frame for the "test" data and vice-versa.
    It's better to split time based data on actual date boundaries so that you are making sure that your test set has no data points that overlap into the training data points and vice-versa.
    What I do is take the first 80% of the top data rows as the "training" set and then everything after as the "test" set. By doing this the results are more accurate.
    (I see a lot of videos use the "split" function to split data and there are a lot of articles about why you shouldn't do this using time based data)
    The other suggestion I would make is to use the XGBoost library as it's better suited for time based predictions.

    • @ecoengineering7535
      @ecoengineering7535  7 месяцев назад

      Thank you very much for expressing your great idea.

  • @michaelalcala2412
    @michaelalcala2412 7 месяцев назад

    Hello! Are you accepting projects like making an EA? I want to learn how to make but for this time I really need someone who is experienced in making this project... thanks

    • @ecoengineering7535
      @ecoengineering7535  7 месяцев назад

      hi, Sorry, no

    • @michaelalcala2412
      @michaelalcala2412 7 месяцев назад

      @@ecoengineering7535 Thank you very much. I watched your videos and learned a lot from you...

  • @akbarkhakpour20akbar42
    @akbarkhakpour20akbar42 7 месяцев назад

    سلام لطفا زبان خودمان هم قرار بدین

  • @SamanWorldExplore
    @SamanWorldExplore 7 месяцев назад

    👌👌👌👍👍👍

  • @narminzeynali7042
    @narminzeynali7042 7 месяцев назад

    👌👌👍👍