How would you challenge this algo?

Discussion in 'Automated Trading' started by Drawdown Addict, Aug 5, 2023.

  1. Hello Drawdown Addict,

    You need more than 50 back test trades my friend.
     
    #51     Aug 7, 2023
  2. Let's post something useful in between all of this garbage.

    This is the response from ChatGPT for a Geometric Brownian Motion simulator in Python. It is pretty neat.

    Code:
    
    import numpy as np
    import matplotlib.pyplot as plt
    
    # Parameters
    mu = 0.1        # Drift (average return)
    sigma = 0.2     # Volatility (standard deviation of return)
    S0 = 100        # Initial stock price
    T = 1           # Time period in years
    N = 252         # Number of trading days in a year (assuming 252)
    dt = T / N      # Time interval
    num_simulations = 5
    
    # Simulate GBM
    np.random.seed(42)
    for _ in range(num_simulations):
        t = np.linspace(0, T, N+1)
        W = np.random.standard_normal(size=N+1)  # Brownian motion increments
        W = np.cumsum(W) * np.sqrt(dt)  # Brownian motion
    
        # Geometric Brownian Motion equation
        S = S0 * np.exp((mu - 0.5 * sigma**2) * t + sigma * W)
      
        # Plot
        plt.plot(t, S, lw=2, alpha=0.6)
    
    plt.xlabel('Time')
    plt.ylabel('Stock Price')
    plt.title('Geometric Brownian Motion Simulation')
    plt.grid(True)
    plt.show()
    
    
     
    #52     Aug 7, 2023
    NASRI and SimpleMeLike like this.
  3. Hello Drawdown Addict,

    Click back test and Post the performance report only.
     
    Last edited: Aug 7, 2023
    #53     Aug 7, 2023
  4. virtusa

    virtusa

    I developped my system on a few months of data (intraday trading). I tested it, so let it run without changing anything, on about 1,000 trades. This represented at that time about 2 years of data. It was clear that in any market circumstances the system always was profitable. Never a single losing week. So that test confirmed that the system did adapt itself very well to any type of market behavior. I never believed , and still don't believe, in computer nerds that write blackboxes, let it do a lot of number crunching and then watch what the result is. These people in general never really studied the markets. I meet several of them.

    To have a dynamical system you should have a profound knowledge of the market behavior as you have to tell your computer what to do and why. ChatGPT is a product of the laziness of people.
    A computer only knows what people feed him. And as most of these computer nerds have no clue about the market they give worthless feed which results in worthless results.
     
    Last edited: Aug 7, 2023
    #54     Aug 7, 2023