This question is related to this one:
How to determine seasonality of a binary variable?
I have a multivariate binary time series. That is, say I have $y_1 \in$ {0, 1}$^N$, $y_2 \in$ {0, 1}$^N$
I expect that $y_1$ and $y_2$ are correlated.
In the linked question, there is a recommendation to use polynomial smoothing (and the OP has only one binary time series). But I don't understand why polynomial smoothing would be helpful.
Question: How can I determine the seasonality of each of the time series independently, and together?
