Is there a wait function?

[pascalchristian]

Is there a wait function for closing/reversing a position? For example entering into a short position (based on criterias) then wait X trading days, then close the position? Thanks

[eervin123]

I'll be very interested in seeing a better answer here, I solved it in what is probably the most ugly manner.

class example(Strategy):
    n1 = 1
    n2 = 20
    relax = 5  # how many days/periods to wait before trading again
    relax = pd.to_timedelta(relax, unit="D")  # convert to a timedelta object
    def init(self):
        super().init()
        self.sma1 = self.I(SMA, self.data.Close, self.n1)
        self.sma2 = self.I(SMA, self.data.Close, self.n2)

    def next(self):
        super().next()
        price = self.data.Close[-1]
        if len(self.closed_trades) == 0: # If there hasn't been any trades yet
            last_trade_time = self.data.index[0]
        elif len(self.trades) == 0: # We are in cash no open trades
            last_trade_time = self.closed_trades[-1].exit_time # When did we exit our last trade
        else: # We have an open position
            last_trade_time = max(
                self.trades[-1].entry_time, self.closed_trades[-1].exit_time
            ) # the max of when we either entered that trade or exited it last
        current_time = self.data.index[-1] 
        dayssince = (
            current_time - last_trade_time
        )  # this will be used in concert with relax variable
        # Bullish Markets ---------------------------------
        if (
            not self.position.is_long
            and crossover(self.sma1, self.sma2)
        ):
            self.buy(size=0.9)
        elif (
            self.position.is_long
            and crossover(self.sma1, self.sma2)
            and dayssince > self.relax # Checks to make sure it has been at least the proper amount of days since last trade
        ):
        ...rest of code here....

[eervin123]

btw there is a barsscince function in the lib.py module but I couldn't quite figure it out so I want with the above ugly method. I'm sure @kernc will be once again rolling his eyes at my bastardization of his beautiful library. ;)

def barssince(condition: Sequence[bool], default=np.inf) -> int:
    """
    Return the number of bars since `condition` sequence was last `True`,
    or if never, return `default`.
        >>> barssince(self.data.Close > self.data.Open)
        3
    """
    return next(compress(range(len(condition)), reversed(condition)), default)

[kernc]

Untested, something like:

cur_time = self.data.index[-1]
for trade in self.trades:
    if barssince(cur_time < trade.entry_time) >= 5:
        trade.close()

I guess it's not its best use-case. 😅

[kernc]

How about:

import pandas as pd

...

    def next(self):
        cur_time = self.data.index[-1]
        for trade in self.trades:
            if cur_time - trade.entry_time >= pd.Timedelta('5 days'):
                trade.close()

[kernc]

I had an idea for a composable base strategy HoldNBarsStrategy that does exactly above but didn't yet manage to come round to forging it.

[eervin123]

I'd vote for that because I still haven't figured out to get mine to work with the optimizer without even more coding.