A time series (seasonal ARIMA) model for varying variability in different seasons
Is there a time series ARIMA model that can represents increased variation (expand confidence intervals) in values during one season? As an example, attached is a data set showing the number of deaths in the United States. A time series analysis of deaths prior to March of 2020 (settings below) gives a reasonable prediction of deaths, however its accuracy is reduced during Dec - Feb (flu season...