Algorithmic Trading Video Series #6 | Indicators + History Requests

hi there,

can i ask you a question about the code, please?

why we are putting this code: if not self.Portfolio[self.spy].IsLong: ？

i put this line in the comment and backtested it. there are some small amount orders which i dont understand.

do you have any method that I can find out how it happen please?

do appreciate.

as i can tell from the orders, the orders were not fully filled in one go sometimes.

but i don't know how it can happen.

Hi Zl cao,

In the original algorithm, we only go long if the given conditions are met, along with we are not already long on the security. We get a few extra orders by commenting out the line as the algorithm will try to buy more shares even if we are long. We can see that in extra orders, the order quantity is very small(1-2); this is because we already have the maximum position possible, but due to a decrease in price, we can add a few more shares to our holdings.

Best,

Varad Kabade

Hello Louis,
I'm trying to get the angle of the SMA with the same approach you used to define your custom indicator.
I'm getting an error when I try to make use the np.diff() method becuase the SMA_values looks to be empty
raise ValueError("diff requires input that is at least one dimensional")
Can you suggest somthing to fix this issue?

class CustomSimpleMovingAverage_angle(PythonIndicator):
    
    def __init__(self, name, period):
        self.Name = name
        self.Time = datetime.min
        self.Value = 0
        self.queue = deque(maxlen=period)
        self.SMA_Values = deque(maxlen=period)

    def Update(self, input):
        self.queue.appendleft(input.Close)
        count = len(self.queue)
        self.SMA_Values.appendleft(sum(self.queue) / count)
        self.Time = input.EndTime
        
        self.Value = sum(self.queue) / count
        
        if count == self.queue.maxlen:
            a = np.diff(self.SMA_Values.reverse())
            self.Value = [(math.degrees(math.atan(x))) for x in a][-1]
        # returns true if ready
        return (count == self.queue.maxlen)

Hi Joseph46

deque.reverse() method returns NoneType object instead of the reversed queue (see here), and it will change the original queue instance. You may consider using the following instead:

a = np.asarray(self.SMA_Values)[::-1]

Best
Louis