Bagging or bootstrap aggregation seems to make sense for time series forecasting using an ensemble because bagging randomizes subsets of the data with replacement. However, I've only seen bagging used for homogeneous base learners when constructing ensembles.
Stacking is another ensemble technique that uses heterogeneous base learners, but stacking employs cross-validation, which I don't view as being appropriate for economic time series forecasting, even if time series split cross-validation that retains the ordering of observations is used.
As you can see, for time series forecasting, i like the random re-sampling of bagging, and the heterogeneity of base algorithms in stacking. How to combine these qualities in ensemble learning?
Can heterogeneous algorithms be used in a bagging ensemble? Or can bootstrap re-sampling be used within stacking?