Kalman Filter in non-linear models

[pfjulio]

Hi Jaromir,

Is is possible to get the smoothed variables (filter the model with the KalmanFilter) in a non-linear model with the left and right hand sides of all equations being exatly equal in each and every period (instead of having smoothed variables based on a first-order approximation of the model, which generates discrepancies between the LHS and the RHS)?

Many thanks

[jaromir-benes]

Hi - no, there is no option to run the kalman filter this way. If you have the same number of shocks as the number of observed variables, you can use nonlinear simulation with an exogenize/endogenize plan. Best, Jaromir

…

On Tue, Oct 11, 2022 at 10:13 AM pfjulio ***@***.***> wrote: Hi Jaromir, Is is possible to get the smoothed variables (filter the model with the KalmanFilter) in a non-linear model with the left and right hand sides of all equations being exatly equal in each and every period (instead of having smoothed variables based on a first-order approximation of the model, which generates discrepancies between the LHS and the RHS)? Many thanks — Reply to this email directly, view it on GitHub <#343>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGCVKKRKZHXFW3FEKZ2RB73WCUOSPANCNFSM6AAAAAARCATRIQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>