# Excess shocks in filtering **URL:** \ **Category:** General DSGE Modeling\ **Created:** [9 July 2026 13:12 UTC](https://forum.dynare.org/t/excess-shocks-in-filtering/28431 "2026-07-09T13:12:21Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Kavarkk](https://forum.dynare.org/letter_avatar_proxy/v4/letter/k/919ad9/32.png) [@Kavarkk](https://forum.dynare.org/u/Kavarkk)\ **Post date:** [9 July 2026 13:12 UTC](https://forum.dynare.org/t/excess-shocks-in-filtering/28431/1 "2026-07-09T13:12:21Z")
Hello, I have a question regarding Dynare’s kalman filter. Is it possible to filter shocks in a model with more shocks than observables. And if it is possible, then how does dynare handle this? Does it use some kind of an optimization-smoother? If understand correctly, it is not possible to identify all shocks in a situation like this ([Excess shocks can limit the economic interpretation - ScienceDirect](https://www.sciencedirect.com/science/article/pii/S0014292122000599?via%3Dihub)), so what kind of procedure does dynare use? thanks in advance ---
**Author:** ![jpfeifer](https://forum.dynare.org/user_avatar/forum.dynare.org/jpfeifer/32/5044_2.png) [@jpfeifer](https://forum.dynare.org/u/jpfeifer)\ **Post date:** [9 July 2026 14:05 UTC](https://forum.dynare.org/t/excess-shocks-in-filtering/28431/2 "2026-07-09T14:05:48Z")
I am not sure I understand the question. The Kalman filter and smoother have unique recursion to compute the filtered and smoothed shocks. The problem is that the original/true structural shocks will typically not cleanly map into recovered smoother and filtered shocks (as described in the paper you referenced). ---
**Author:** ![Kavarkk](https://forum.dynare.org/letter_avatar_proxy/v4/letter/k/919ad9/32.png) [@Kavarkk](https://forum.dynare.org/u/Kavarkk)\ **Post date:** [9 July 2026 15:48 UTC](https://forum.dynare.org/t/excess-shocks-in-filtering/28431/3 "2026-07-09T15:48:48Z")
Sorry, I guess I was confused on the Kalman smoother. Thank you for your answer