Hi,
I've noticed several open issues on GitHub regarding that Quarterly forecast for current quarter using DynamicFactorMQ does not update when new data for the month ending that quarter is added.
But, I haven't seen any indication that these issues have been closed or resolved. Is it still an open limitation of the model?
For example:
One possible manual workaround would be to compute the impact of the news (i.e., the new monthly data that affects the quarterly variable) and add this impact to the forecast obtained using data only up to the second month of the quarter.
Would this approach be valid? It seems like a reasonable way to incorporate the new information when the model itself doesn't update the forecast automatically.