rrr.sim5: Simulation model 5

In rrpack: Reduced-Rank Regression

Simulation model 5

Description

Generate data from a mean-shifted reduced-rank regression model

Usage

rrr.sim5(
  n = 40,
  p = 100,
  q = 50,
  nrank = 5,
  rx = 10,
  s2n = 1,
  rho_X = 0,
  rho_E = 0,
  nout = 10,
  vout = NULL,
  voutsd = 2,
  nlev = 10,
  vlev = 10,
  vlevsd = NULL,
  SigmaX = "CorrCS",
  SigmaE = "CorrCS"
)

Arguments

n	sample size
p	number of predictors
q	numbers of responses
nrank	model rank
rx	rank of the design matrix
s2n	signal to noise ratio
rho_X	correlation parameter for predictors
rho_E	correlation parameter for errors
nout	number of outliers; should be smaller than n
vout	control mean-shifted value of outliers
voutsd	control mean-shifted magnitude of outliers
nlev	number of high-leverage outliers
vlev	control value of leverage
vlevsd	control magnitude of leverage
SigmaX	correlation structure of predictors
SigmaE	correlation structure of errors

Value

similated model and data

References

She, Y. and Chen, K. (2017) Robust reduced-rank regression. Biometrika, 104 (3), 633–647.

rrpack documentation built on June 16, 2022, 9:05 a.m.