I have a model like this: $$P_t=\alpha_t+gP_{t-1}+u_t$$

$$\alpha_t=\alpha_{t-1}+d+n_t$$ Where {$u_t$} and {$n_t$} are normal, iid with 0 mean but unknown variance.

I want to estimate the parameters $g$ and $d$, and all I have is a series of $P_t$. Can I solve this problem by using Kalman Filter? If yes, how to do this with pykalman (or some other python packages); if no, what else can I use?

This problem origins from an empirical study about market quality, you can find the paper here: https://www.sciencedirect.com/science/article/pii/S0378426608000393

  • $\begingroup$ "Can I solve this problem by using Kalman Filter?" <- yes! (depending on what you mean by solve) $\endgroup$ – Taylor Mar 14 at 3:40
  • $\begingroup$ By saying "solve" I mean "get reliable estimations of the parameters" :) $\endgroup$ – jqf Mar 14 at 3:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.