VARMLE: generalized Dantzig selector for transition matrix update in...

In hdiVAR: Statistical Inference for Noisy Vector Autoregression

generalized Dantzig selector for transition matrix update in maximization step

Description

Sparse estimation of transtion matrix in vector autoregression given conditional autocovariance matrices.

Usage

VARMLE(S0, S1, tol)

Arguments

S0	a p by p matrix; average (over time points) of conditional expectation of x_t x_t^\top on y_1, \ldots, y_T and parameter estimates, obtained from expectation step.
S1	a p by p matrix; average (over time points) of conditional expectation of x_t x_{t+1}^\topon y_1, \ldots, y_T and parameter estimates, obtained from expectation step.
tol	tolerance parameter in Dantzig selector.

Value

Sparse estimate of transition matrix by Dantzig selector.

Author(s)

Xiang Lyu, Jian Kang, Lexin Li

hdiVAR documentation built on May 31, 2023, 7:27 p.m.