Is there anyway to save down the Sharp-Ratios and Other Parameters from BackTesting.Optimize

[kiann00]

Hi there,

I am still trying out the backtesting.optimize function (which is GREAT!!!). One think I would like to do, for analysis, though, is to see the various metrics that are pumped out as part of the backtesting run.

For example, I understand the below example shows the 'heatmap' which shows each of the input variables, and the output for the desired maximize function...
BUT, is there a way to save other metrics, such as 'Sharpe Ratio', 'Volatility' etc, that were calculated from a 'straight' run?

Thanks, gurus!

#############
s_time = time.time()
bt = Backtest(data_test, RsiOscillator, cash=10_000, commission=.002)
stats_skopt_2 = bt.run()
# bt.plot()
print('time for the initial strategy test is', time.time() - s_time)

def maximize_func(stats):
    # return stats['Equity Final [$]'] / stats['Max. Drawdown [%]'] * stats['Expectancy [%]']
    if stats['Max. Drawdown [%]'] == 0:
        return stats['Equity Final [$]'] * stats['Expectancy [%]']
    else:
        return stats['Equity Final [$]'] * stats['Expectancy [%]'] / stats['Max. Drawdown [%]']

s_time = time.time()

# Note : for method = 'skopt', we only need interval end-points
# We can input constraints per this line if needed : constraint=lambda p: p.rsi_lower < p.rsi_upper,

stats_skopt_2, heatmap, optimize_result = bt.optimize(
    upper_bound = [60, 70],
    lower_bound = [15, 30],
    rsi_window = [11, 14],
    maximize= maximize_func,
    method='skopt',
    max_tries=100,
    random_state=0,
    return_heatmap=True,
    return_optimization=True)

[kernc]

You could save and/or examine every instance of stats that you get into your maximize_func() ...

[kiann00]

sorry, @kernc , but where would I put the code to save down the stats within the bt.optimize?

[sabs-io]

Did you find an answer to this? I would be interested to know how to save every instance of stats and trades as my optimisations can take hours to complete.

[kernc]

I thought one can save all optimization stats somewhat like so:

all_results_stats = []

def maximize_func(stats):
    all_results_stats.append(stats)
    return stats['SQN']  # or similar

bt = Backtest(...)
bt.optimize(..., maximize=maximize_func)

assert all_results_stats

but now I'm thinking this simple approach may not work with multiprocess optimization. 😅

[kiann00]

@daveysab / @kernc ,

@kernc , thanks for the advice (code)..... I tried your sample, and unfortunately, it did not seem to save down anything. Perhaps I did something wrong (though I just cut-and-paste your suggestion). And yes, the trouble is, I am trying to have multi-process optimization...

@daveysab , unfortunately, I did not manage to debug/get the answer to this as of yet. If I manage to debug it, I will post it here, and give you a shout.