So I understand what a moving average is and how to calculate it. I'm using this numpy function for it.
I am somehow confused about how to calculate rolling returns. Different sites explain it differently, and some conflate the two. After careful digging, however, I think I've understood the following:
n-period Moving Average: is just the average of the last n periods for every period.
n-period Rolling Returns: is just the n-period Moving Average of returns where returns are defined as:
def rolling_returns(positions, n=3): returns = (positions-positions)/positions return moving_average(returns,n)
Where positions is just a vector/numpy array of our position as a function of time. Am I missing/conflating something? I'm not really sure if this is right. Any help would be appreciated