0
$\begingroup$

Statistics gurus,

Kalman filter appears to be a powerful estimator for linear problems. I understand one can tune the performance by adjusting parameters like process noise and measurement noise. Is it possible to adjust these parameters to make Kalman filter results converge to a classic linear regression? If yes, how? Please kindly share your opinions. Thanks.

Rgds,

Improve this question
$\endgroup$
1

1 Answer 1

Reset to default
0
$\begingroup$

Yes, this is possible by setting the process noise to zero, assuming your coefficients (which are in the state) are deterministic. The filtered estimate of the state at the final observation is then equal to the estimate of a classic linear regression, if an exact initialisation is employed.

Improve this answer
$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.