rankest.sdr: Estimate the true number of dimensions for sdr models
In abnormally-distributed/cvreg: Cross Validation and Robust Estimation Utilities

Description Usage Arguments Value References

This offers two different BIC-type statistics for the selection of ranks. The ZMP and LAL variants are offered here (Zhu, Miao, & Peng, 2006; Li, Artemiou, & Li, 2011). Note, however, that despite the BIC-type penalty, these statistics do not include the -2 multiplier. Hence, the largest value is optimal rather than the smallest as in the BIC.

rankest.sdr(
  fit,
  criterion = c("zmp", "lal"),
  max.rank = ncol(fit$mf[, -1]),
  plot = T
)

`fit`	the model fit
`criterion`	one of "zmp" (the default) or "lal"
`max.rank`	the maximum rank to consider. defaults to all of them.
`plot`	should the BIC-curve be plotted? defaults to TRUE.

a list

Zhu, L., Miao, B., & Peng, H. (2006). On Sliced Inverse Regression With High-Dimensional Covariates. Journal of the American Statistical Association, 101(474), 630–643. doi:10.1198/016214505000001285

Li, B., Artemiou, A., & Li, L. (2011). Principal support vector machines for linear and nonlinear sufficient dimension reduction. The Annals of Statistics, 39(6), 3182–3210. doi:10.1214/11-aos932

abnormally-distributed/cvreg documentation built on May 3, 2020, 3:45 p.m.