Prediction intervals

[michael-bag]

Beautiful implementation of dynamic factor models, thanks a lot.

Is it true that we don't (yet) have a way to calculate prediction intervals?

(Barigozzi and Hallin 2019 looks interesting: https://arxiv.org/abs/1811.10045)

[SebKrantz]

Thanks! Truthfully my time on this package is limited. And I work more on transport these days – just releasing a significant library. But, if you can facilitate me is some way, e.g., provide research codes, or, better, fork the library, download the research code into the same folder, open in Cursor AI or related editor, and use a flagship model (Opus 4.5 or GPT 5.2 Codex) to implement (use plan mode for optimal results), and send me a PR. That would dramatically speed up things.