tenAR.sim: Generate TenAR(p) tensor time series

Description Usage Arguments Value See Also Examples

View source: R/tenAR.R

Description

Simulate from the TenAR(p) model.

Usage

1
tenAR.sim(t, dim, R, P, rho, cov, A = NULL, Sig = NULL)

Arguments

t

length of output series, a strictly positive integer.

dim

dimension of the tensor at each time.

R

Kronecker rank for each lag.

P

autoregressive order.

rho

spectral radius of coefficient matrix Φ.

cov

covariance matrix of the error term: diagonal ("iid"), separable ("mle"), random ("svd").

A

coefficient matrices. If not provided, they are randomly generated according to given dim, R, P and rho. It is a multi-layer list, the first layer for the lag 1 ≤ i ≤ P, the second the term 1 ≤ r ≤ R, and the third the mode 1 ≤ k ≤ K. See "Details" of tenAR.est.

Sig

only if cov=mle, a list of initial values of Σ_1,…,Σ_K. The default are identity matrices.

Value

A tensor-valued time series generated by the TenAR(p) model.

See Also

tenFM.sim

Examples

1
2
3
set.seed(123)
dim <- c(3,3,3)
xx <- tenAR.sim(t=500, dim, R=2, P=1, rho=0.5, cov='iid')

tensorTS documentation built on Aug. 10, 2021, 9:07 a.m.