Finding Top n losing trade in backtest result

i think you can download the backtest json directly from quant connect cloud. or if you run it locally it will be generated. order csv can be download from cloud backtest as well. not sure which has pnl per trade

Thx for reply, indeed, i am looking for the way to calculate pnl from trade 

eg 

i buy apple at 20230101 100 share $150

and then i sell

apple at 20230102 100 share $151

i want to know i earned $100 from this trade

below is some i can download

20230101,appl,100,150

20230102,appl,-100,151

but i want to find out i earned $100 from 20230101 purchase of appl

i knew that the code is not so complicated. but I don’t want to reinvent the wheel . hope that there is reuseable libraries or code for the purpose 

Hey Enjolras Leigh

you could make a data frame adding the symbols and the last made profit and then import that dataframe via objectstore to your notebook.

def OnOrderEvent(self, orderEvent):
            order = self.Transactions.GetOrderById(orderEvent.OrderId)
            if order.Quantity < 0:
                data = {
                        'symbol':order.Symbol.Value,
                        'time':self.Time,
                        'quantity': order.AbsoluteQuantity,
                        'profit': self.Portfolio[order.Symbol].LastTradeProfit
                    }
                if self.orders_dataframe is None:
                    self.orders_dataframe = pd.DataFrame(data=data,index=[0])
                else:
                    self.orders_dataframe = self.orders_dataframe.append(data, ignore_index=True)

This will only work if you buy into the up direction and sell to the down direction. If you also short stocks you might want to change the if statement. I also added time and quantity maybe it might help you.

At the end you have to add this:

def OnEndOfAlgorithm(self):
           
            self.ObjectStore.Save(f"{self.ProjectId}/bytes_key", self.profit_dataframe.to_json())

and in the notebook you will have to access it like this:

profit_df = pd.read_json(qb.ObjectStore.Read(f"{qb.ProjectId}/bytes_key"))
profit_df.time = pd.to_datetime(profit_df.time, unit='ms')

Hope it helps!

Best, 

Nico