Generate Simulation Data for Benchmarking Sparse Regressions (Cox Model)

Generate simulation data for benchmarking sparse Cox regression models.

Usage

msaenet.sim.cox(
  n = 300,
  p = 500,
  rho = 0.5,
  coef = rep(0.2, 50),
  snr = 1,
  p.train = 0.7,
  seed = 1001
)

Arguments

n: Number of observations.
p: Number of variables.
rho: Correlation base for generating correlated variables.
coef: Vector of non-zero coefficients.
snr: Signal-to-noise ratio (SNR).
p.train: Percentage of training set.
seed: Random seed for reproducibility.

Value

List of x.tr, x.te, y.tr, and y.te.

References

Simon, N., Friedman, J., Hastie, T., & Tibshirani, R. (2011). Regularization Paths for Cox's Proportional Hazards Model via Coordinate Descent. Journal of Statistical Software, 39(5), 1--13.

Author

Nan Xiao <https://nanx.me>

Examples

dat <- msaenet.sim.cox(
  n = 300, p = 500, rho = 0.6,
  coef = rep(1, 10), snr = 3, p.train = 0.7,
  seed = 1001
)

dim(dat$x.tr)
#> [1] 210 500
dim(dat$x.te)
#> [1]  90 500
dim(dat$y.tr)
#> [1] 210   2
dim(dat$y.te)
#> [1] 90  2