Posts made by marketmaker

@yourdatatalking
Hi man.. well, I'm asking very basic questions (I'm beginner) like how to program ema21 change of direction, but nobody is answering me.

so there are 2 possibilities:

1. algotrading is bullshit and everybody quit months ago
   or
2. algotrading is earning billions of dollars and nobody will help you because it is you vs. them

now the billion dollar question is - which answer is the right one :)

hi guys,

I'm using backtrader and just realized that it works nicely but sometimes, there is this situation:

the strategy will execute last buy but then, there is no last sell -> so it is hard to tell how successful was strategy in PNL.

is there any parameter or how to just compare starting cash with cash on end after last sell? so if after last sell, there is another one buy (but no other sell)-> to ignore that buy?

something like cheat on end or?

example:
start cash was = 250 000
last sell: 10 BTC for 50000 => 500 000 cash.
gain is 100% (500 000 vs 250 000)
but if there is another buy after last sell, like for 400 000, how to calculate PNL correctly?

hi guys,
I would like to have a buy signal when ema21 change it's direction (stop falling, starts rising). is there any quick way to do it or do I need to access previous day ema21 to compare values if there is a direction change?

I have this but it doesn't work as expected:

ema21 = bt.ind.EMA(period=21)
self.ema21up = bt.indicators.UpDay(ema21)

thank you...

@marketmaker I'm an idiot :))

correct if is this:
if self.params.temp * 1.05 < self.data.close[0] :

then it works :)

btw any better way to find out buy price as my approach
self.params.temp = self.data.close[0]
?

hi guys,
first of all - I cannot even imagine that software like backtrader is free - so much work and development like - big thank to creator!

anyway, I'm beginner in python and playing with backtrader and I have working code, but it doesn't work :) cannot find why -> I want to SELL when close price is lower than was buy price + 5 percent.

my code is very simple and temp is used to store buy price:

    class TestStrategy(bt.Strategy):

        params = ( ('pfast', 20), ('pslow', 50), ('order_percentage', 0.95), ('temp', 0.00))

        def __init__(self):
            sma1, sma2 = bt.ind.SMA(period=self.p.pfast), bt.ind.SMA(period=self.p.pslow)
            self.crossover = bt.indicators.CrossOver(sma1,sma2)

        def next(self):
            #print(self.data.close[0])    #last close price
            if self.position.size == 0:
                if self.crossover > 0:
                    amount_to_invest = (self.params.order_percentage * self.broker.cash)
                    self.size = math.floor(amount_to_invest / self.data.close)
                    print(f"Buy {self.size} for {self.data.close[0]} USD")
                    self.params.temp = self.data.close[0]
                    self.buy(size=self.size)


            if self.position.size > 0:
#HERE IS UNKNOWN BUG - it doesn't sell with 5% profit
                if self.params.temp < self.data.close[0] * 1.05:
                    self.close()
                    print(f"SELL {self.size} for {self.data.close[0]} USD")

on apple data it looks like this:

Buy 129 for 73.45 USD
SELL 129 for 72.2675 USD
Buy 80 for 119.03 USD
SELL 80 for 118.69 USD

so it sell for not +5% but even cheaper -> what is wrong? can you help me?

and also, is there any better way to store buy price as I'm doing? (I believe my approach is not very clean code..)

thank you in advance