retirement_portfolio_5.py

def get_simulation(data):
    # Get the logarithmic returns of the % change of prices from one trading day to the next.
    log_returns = np.log(1 + data.pct_change()) 
    # Get the mean of these returns
    u = log_returns.mean() 
    # Get the variance of these returns 
    var = log_returns.var()
    # Get the change in the average value of these values
    drift = u - (0.5 * var)
    # Get the standard deviation
    stdev = log_returns.std()
    # How many days are we going to run the stimulation for
    t_intervals = DAYS_IN_YEAR * YEARS
    # How many simulations of this financial instrument are we going to run? 
    iterations = 10
    # Create the Monte Carlo stimulation of daily percent changes of the financial instruments.
    daily_returns = np.exp(drift + stdev * norm.ppf(np.random.rand(t_intervals, iterations)))
    # Create an numpy array filled with zeros with the same shape as the daily_returns numpy array.
    price_list = np.zeros_like(daily_returns)
    # Set the most recent trading day's data as the start prices
    price_list[0] = Sdata.iloc[-1]
    # For each day in the simulation, compute the price of the stock after multiplying 
    # the previous's price by the current day's price.
    for t in range(1, t_intervals):
        price_list[t] = price_list[t - 1] * daily_returns[t]

    # Get all the percentage returns for all the simulations for this financial instructment.
    asset_returns = price_list[-1] / price_list[0]
    # Get and return the geometric mean (because we are dealingn with percentages)
    # of all these simulations for this financial instrument.                     
    return gmean(asset_returns)

Hi Dirk,

Thanks for you comments. I'd missed those errors when I was copying and pasting the code to create the gists.

Thanks again

Shane