Closed olshena closed 4 years ago
The both have seeds in them. I don't know that including the seed as an argument is functionally different than just leaving the user to set it before calling the function, but I think I like making it an explicit argument.
predictAR has a seed built in, while simulateHoltMA does not. We should standardize in one way. Which is better?