Accessing stats["_trades"] during the test

[MaxvanderWerf]

I'm trying to code a variation of Dollar Cost Averaging and therefore need to know the trades that have already been made during the test to determine if I should buy or not. Initializing self.stats = self.stats["_trades"] in the strategy gives me an error. Does anyone know if this is possible at all?

class DCA_every_week(Strategy):

    amount_to_invest = 10000

    def init(self):
        
        self.day_of_week = self.I(
                            lambda x: x,
                            self.data.Close.s.index.dayofweek,
                            plot = False,
                            )
        
        self.stats = self.stats["_trades"]

    def next(self):

        if self.day_of_week[-1] == 4: 
            self.buy(size = math.floor(self.amount_to_invest / self.data.Close[-1])) # Buys largest number of stocks possible

AttributeError: 'DCA_every_week' object has no attribute 'stats'

[kernc]

There is no Strategy.stats! stats are returned by Backtest.run().
During the run, you can maybe look in Strategy.closed_trades?

[MaxvanderWerf]

Thanks! If anyone else is wondering, I found them using Strategy.trades (with DCA you only buy and don't sell).

[MaxvanderWerf]

One more question:

I can get the length of the list of active trades through len(self.trades), however I cannot access the variables of trades.

When I try:
self.trades.entry_time
I get:
AttributeError: 'tuple' object has no attribute 'entry_time'

When I try to get the latest trade by using:
self.trades[0].entry_time
I get:
IndexError: tuple index out of range

Does anyone know how to access the instance variables of a trade?

[kernc]

When I try to get the latest trade by using: self.trades[0].entry_time I get: IndexError: tuple index out of range

At the beginning when there are no trades yet, 0 in trades[0] is an invalid index. Hide it behind if self.trades: check.