I have this time series of seasonal monthly data, which sometimes has missing observations. The likelihood of an observation being missing is mostly dependent on the month, but also of the value had it not been missing. I have an estimate for the value of the series in the next observation. And I want to estimate whether it will actually be missing from the series (i.e. non-observed). Obviously, I also know what month it will be. I can make estimates for the missing (unobserved) data in the past (SARIMA and\or external indicator). What's the best way to approach this?
