Strategy optimization and benchmarking

Financial Trading in Python

Chelsea Yang

Data Science Instructor

How to decide the values of input parameters?

Question:
- Which is a better SMA lookback period to use in the signal "Price > SMA"?
Solution: strategy optimization
- Try a range of input parameter values in backtesting and compare the results

Strategy optimization example

def signal_strategy(ticker, period, name, start='2018-4-1', end='2020-11-1'):

    # Get the data and calculate SMA
    price_data = bt.get(ticker, start=start, end=end)
    sma = price_data.rolling(period).mean()

    # Define the signal-based strategy
    bt_strategy = bt.Strategy(name, 
                              [bt.algos.SelectWhere(price_data>sma),
                               bt.algos.WeighEqually(),
                               bt.algos.Rebalance()])

    # Return the backtest
    return bt.Backtest(bt_strategy, price_data)

¹ www.datacamp.com/courses/writing-functions-in-python

Strategy optimization example

ticker = 'aapl'
sma20 = signal_strategy(ticker, 
                        period=20, name='SMA20')
sma50 = signal_strategy(ticker, 
                        period=50, name='SMA50')
sma100 = signal_strategy(ticker, 
                        period=100, name='SMA100')


# Run backtests and compare results
bt_results = bt.run(sma20, sma50, sma100)
bt_results.plot(title='Strategy optimization')

A plot of the optimization result

What is a benchmark?

A standard or point of reference against which a strategy can be compared or assessed.

Example:
- A strategy that uses signals to actively trade stocks can use a passive buy and hold strategy as a benchmark.
- The S&P 500 index is often used as a benchmark for equities.
- U.S. Treasuries are used for measuring bond risks and returns.

Benchmarking example

def buy_and_hold(ticker, name, start='2018-11-1', end='2020-12-1'):

    # Get the data
    price_data = bt.get(ticker, start=start_date, end=end_date)

    # Define the benchmark strategy
    bt_strategy = bt.Strategy(name, 
                              [bt.algos.RunOnce(),
                               bt.algos.SelectAll(),
                               bt.algos.WeighEqually(),
                               bt.algos.Rebalance()])

    # Return the backtest
    return bt.Backtest(bt_strategy, price_data)

Benchmarking example

benchmark = buy_and_hold(ticker, name='benchmark')

# Run all backtests and plot the resutls
bt_results = bt.run(sma20, sma50, sma100, benchmark)
bt_results.plot(title='Strategy benchmarking')

Benchmarking result

Let's practice!

Financial Trading in Python