xreg with predicted values #153
Description
There should be a mechanism of defining, which variables should be predicted and which should be used as is. A method preparing explanatory variables is needed.
When the variable is marked for prediction this is what should happen inside the function:
- Predict each marked explanatory variable;
- Use fitted values for the estimation;
- Use point forecasts for forecasting;
- Use covariances between variables for the variance calculation;
- Use (3) and (4) for the final forecasts.
The related issue: how to make this work with pure multiplicative and mixed models.