Prior precision for VAR(1) on K time points (d-dimensional)

Model: U_1 ~ N(0, P0^-1), and for k>=2: U_k | U_k-1 ~ N(A U_k-1, Q) Returns the joint precision matrix for vec(U_1,...,U_K).

make_VAR1_precision(K, d, A = NULL, Q = NULL, P0 = NULL)

Arguments

K

integer number of time points.

d

integer state dimension.

A

d x d transition matrix (default 0.8 * I_d).

Q

d x d innovation covariance (default 0.5 * I_d).

P0

d x d initial precision (default I_d).

Value

(dK) x (dK) precision matrix (dense).