Problem fetching execution price

stanski

Hello everyone,

I'm trying to exit a position if price > executed price + x.
How can I fetch the order.executed.price?

I tried it with _orders[data][0]but it doesn't work:

def notify_order(self, order):
        if order.status in [order.Submitted, order.Accepted]:
            return
        if order.status in [order.Completed]:
            if order.isbuy():
                
                DISTCALC = ((self.ema1[0]-order.executed.price)/order.executed.price*100)
                self.log('BUY EXECUTED, %.2f' % order.executed.price)
                self.log('EMA Price, %.2f' % self.ema1[0])
                self.log('Price distance to EMA, %.2f' % DISTCALC + str('%'))
                
            elif order.issell():
                self.log('SELL EXECUTED, %.2f' % order.executed.price)
           
    def next(self):
        if not self.position and self.order==None: 
            if self.distance and self.below800EMA:
                self.price = self.sma1[0]
                self.order = self.buy(self.datas[0], exectype=bt.Order.Market, size=1)
                  
            if self.sma1[0] > self._orders[data][0]+10:
                
                self.order = self.close()

The error is:
TypeError: list indices must be integers or slices, not GenericCSVData

vladisld

@stanski _orders is designed to be only used internally in Backtrader framework I think.

The better way to implement what you want is to probably manage an association (dictionary) between data and order (assuming only a single active order could ever be created for each data) in your own strategy class.

run-out

@stanski For a non-standard, non-approved and untested answer (well, I use it all the time... )

When you are in notify order that your new order is complete and your position is filled, and preferrably when no one is looking....

Attach an attribute to your data line that is the value of the executed price.

# In notify order and presuming a long only strategy for simplicity.
if order.status in [order.Completed]:    
    if order.isbuy():    
        order.data.enter_price = order.executed.price

Then in your next:

if self.datas[0].close[0] > self.datas[0].enter_price + x:
    <more code>

stanski

@run-out I've tried it with the attached attribute, but unfortunately, I get an AttributeError: 'Lines_LineSeries_DataSeries_OHLC_OHLCDateTime_Abst' object has no attribute 'enter_price

run-out

@stanski That's why I should have tested! Missed a couple lines.

We have to initialize the variable on the datas in init. And then use if to check if None in next.

def __init__(self):
    self.datas[0].enter_price = None

...
if self.datas[0].enter_price:
    print(f"From next: {self.datas[0].enter_price}")

Output
From next: 3239.5
From next: 3239.5
From next: 3239.5
From next: 3239.5
From next: 3239.5

hghhgghdf dfdf

just use self.position.price.

self.positions[data].price if using multiple data sets.

stanski

@run-out well, whatever I do now with self.datas[0].enter_price in terms of multiplication or addition, I get a TypeError: unsupported operand type(s) for *: 'float' and 'NoneType'
I'm considering to discard this idea. This is actually an alternative try-out for a percentage-based take profit, which I couldn't get working in another way.
Maybe I should post the initial code in a new post

run-out

@stanski No worries. It's just a technique I use from time to time when attaching info on datas.

stanski

@run-out nonetheless, thank you for sharing the idea! I just made a new post with the initial code with my take profit problem

alessandro zuin

in the:

def next(self) 
...
    if self.position:
        print(self.position.price)

it works for me, also in a parallel environment.