26 Comments

  1. Lets say you can traid at 3/4 percent accuracy and i can trade at 3/4 percent accuracy if we combine our results we can achieve 7/8 percent accuracy simply by if we agree on wither its a buy or sell we take action but if we disagree neither of us take action. Lets say each of us pick a random number between 1 and 4 whats the chances of us both picking a random number and both of us picking a 4 its 1/8 percent of failure but lets say theres more people trading each additional traider increases the chances of success no need for a machine learning ai to beat the system all we need is basic math

  2. Python, I am very new to ML/Python so I was looking for some help. Can you tell me what kind of setup I would need to optimally trade 2 or so pairs of currency? I am looking to build a new machine and the setup I had in mind was as follows: AMD R9 3900x, Gigabyte Aorus Master X570, 1 TB NVME, 64GB DDR4 3600, 850W PSU( to start), RTX 2080ti (single to begin then adding a second in NVLINK). Both CPU and GPU will be cooled in a custom loop. Thank you for your help.

  3. If you are going through these videos be sure to use an older version of plotly. I rolled back to 3.3.0 in order to make sure the tests were the same as the examples. pip install plotly==3.3.0

  4. Great tutorial. Thanks for using the ideas from my paper. By the way, I have updated the algorithm and have the updated code in Matlab and now using more classifiers like XGboost and logistic regression.

  5. I've been trying to implement my scaloing strategy into code for a while. My eyes and my brain can recognize the charts and what's going to happen. Turning that into code is hard.

  6. Price action is a fundamental piece of algotrading, wether is an indicator, RSI, volume, to name a few, which the vast mayority do A KNOWLEDGE what price is Doing.
    Been trading for a while, and the only true indicator to predict price, is price itself.
    If you think about it, Bollinger bands, MACD, stochastics, Fourier, sine and what not, ARE ALL NOTHING MORE THEN MATHEMATICAL DERIVATIVES OF PRICE ACTION.

  7. How long did it take you to learn this? I don't have much of a coding background but for my model in graduate school I need a daily bias predictor, your original concept from the paper is daily. Thanks for showing us all the framework of building it

  8. Hi buddy, great playlist you got here, I am big ML fan in Financial Markets.
    I am video 2 and I get this error:

    Traceback (most recent call last):
    File "C:/Users/Jay Son/OneDrive/HIT400/MLforecasting/data processing.py", line 8, in <module>
    df.date = pd.to_datetime(df.date,format='%d.%m.%Y %H:%M%:S.%f')
    File "C:UsersJay SonOneDriveHIT400MLforecastingvenvlibsite-packagespandascoretoolsdatetimes.py", line 376, in to_datetime
    result = _assemble_from_unit_mappings(arg, errors=errors)
    File "C:UsersJay SonOneDriveHIT400MLforecastingvenvlibsite-packagespandascoretoolsdatetimes.py", line 446, in _assemble_from_unit_mappings
    unit = {k: f(k) for k in arg.keys()}
    File "C:UsersJay SonOneDriveHIT400MLforecastingvenvlibsite-packagespandascoretoolsdatetimes.py", line 446, in <dictcomp>
    unit = {k: f(k) for k in arg.keys()}
    File "C:UsersJay SonOneDriveHIT400MLforecastingvenvlibsite-packagespandascoretoolsdatetimes.py", line 441, in f
    if value.lower() in _unit_map:
    AttributeError: 'tuple' object has no attribute 'lower'

    Any help?

  9. I am Areej Baasher one of authors of the Research Paper "Forex Daily Trend Prediction using Machine Learning Techniques"
    which you used as a resource, you did a good job youtubing this tutorial.
    Where did you reach in applying the algorithm?

  10. I am Areej Baasher one of authors of the Research Paper "Forex Daily Trend Prediction using Machine Learning Techniques"
    which you used as a resource, you did a good job youtubing this tutorial.
    Where did you reach in applying the algorithm?

Leave a Reply

Your email address will not be published.


*