Backtesting using closing price instead of next day open?

To implement a strategy that places a trade at the close on the same day using the closing price, you can use the market_on_close_order method in QuantConnect. Here's how you can set up your algorithm:

1. Initialize the Algorithm: Set the start and end dates, initial cash, and add the desired equity to your algorithm.

2. Set Data Resolution: Use daily bars for your data resolution to align with your strategy's requirements.

3. Define Market On Close Order: Utilize the market_on_close_order method to place orders at the market close.

4. Calculate Conditions: In the on_data method, calculate if today's close is at least 1% below yesterday's close.

5. Place Order: If the condition is met, use self.market_on_close_order("SYMBOL", quantity) to place the order at the market close.

6. Adjust Submission Time Buffer: If necessary, adjust the MarketOnCloseOrder.submission_time_buffer to ensure the order is submitted within the required time frame before market close.

Here's a sample implementation:

from datetime import timedelta
class MarketOnCloseOrderAlgorithm(QCAlgorithm):
    def initialize(self):
        self.set_start_date(2021, 1, 1)
        self.set_end_date(2021, 12, 31)
        self.set_cash(100000)
        self.add_equity("SPY", Resolution.DAILY)
        self.previous_close = None
        # Adjust the submission time buffer if needed
        MarketOnCloseOrder.SUBMISSION_TIME_BUFFER = timedelta(minutes=10)
    def on_data(self, data):
        if not data.bars.contains_key("SPY"):
            return
        current_close = data["SPY"].close
        if self.previous_close is not None:
            # Check if today's close is at least 1% below yesterday's close
            if current_close <= self.previous_close * 0.99:
                # Place a market on close order
                self.market_on_close_order("SPY", 10)
        # Update the previous close
        self.previous_close = current_close

This code will place a market on close order if today's close is at least 1% below yesterday's close. Make sure to adjust the MarketOnCloseOrder.submission_time_buffer if your exchange allows for a different submission time.

Sources: Market On Open On Close Algorithm, Market On Close Orders - QuantConnect.com, Market On Close Order Buffer Regression Algorithm, Market Orders - QuantConnect.com, Intraday ETF Momentum