TensEnv_dim: Envelope dimension selection for tensor response regression

In kusakehan/TEReg: An R package for tensor envelope regression with tensor response and tensor predictor

Description

This function use the 1D-BIC criterion by by Zhang, X., & Mai, Q. (2018) to select envelope dimensions in tensor response regression.

Usage

TensEnv_dim(Yn, Xn, multiD=1, bic_max=10, opts=NULL)

Arguments

Xn	A vector predictor of dimension p.
Yn	The response tensor instance r_1\times \cdots \times r_m.
multiD	The parameter in ballGBB1D_bic.
bic_max	The maximum envelope dimension to be considered, the parameter in ballGBB1D_bic.
opts	The parameter in ballGBB1D_bic.

Value

u	The envelope dimension of (u_1, u_2,\cdots,u_m).

kusakehan/TEReg documentation built on May 30, 2019, 7:17 a.m.