I am trying to fit the following state space model.
(1) Kt = K(t-1)* + ε1t
(2) Yt = Kt + βZt + ε2t
where, t is time, Yt is the observable variable (at t), Kt is the unobservable trend, and Zt is a matrix of observable variables that might explain Yt. ε1 and ε2 are the usual error terms in a state space model.
I need to estimate the following process by MLE, and get the trend Kt, and the matrix of coefficients β. However I did not find the way to do it with, dlm, nor dlmModReg.
I know how to estimate this model in Matlab (see the link below). But I do not see the way to specify a model like this one with the dlm package. Is it possible to do it with the functions of the dlm package?
(I saw that there is an unanswered related post the title "Exogenous variables in dlm package")
Any help will be sincerely appreciated (even if it implies using another package)!